Brine Seep Microbial Observatory

Examining Microbial Abundance, Diversity, Associations, and Activity at Seafloor Brine Seeps

This project is a collaborative effort between the Joye group, Dr. Ian MacDonald (Florida State University), Dr. Andreas Teske (University of North Carolina), Dr. Kai Uwe-Hinrichs (University of Bremen Germany), and Drs. Kirsten Habicht, Bo Thamdrup and Don Canfield (University of Southern Denmark). We are working at several brine seep sites in the Gulf of Mexico, which are described below in more detail. We conducted a cruise in November and December 2010 and have additional cruises in 2012 and 2013.

Why study Seafloor Brines?

Brine_Fig1_0.jpgBrine seeps occur along the seafloor at sites across the globe. In the northern Gulf of Mexico (GoM), sediments overlie enormous reservoirs of liquid and gaseous hydrocarbons that rest upon Jurassic-age salt deposits. Movement of salt bodies, often referred to as salt-tectonics, creates fault networks that serve as conduits for the rapid transfer of oil, gas and brines from deep reservoirs through the sediments to the seafloor (Fig. 1). Brine expulsion generates brine pools, brine-filled basins, and mud volcanoes.


We are examining the microbiology of two types of brine seeps in the GoM. Brine pools (BP) occur where brine fills surface depressions creating “lake-like” features on the seafloor. Mud volcanoes (MV) occur where brine, oil, gas, and fluidized mud are actively expelled, inducing large temperature fluctuations and seabed alterations. (Fig. 1A)                                             


The chemical composition and salinity of brines influences the composition and activity of the resident microbial community. To date, most of what we know about the microbiology of seafloor brines comes from studies in the eastern Mediterranean (Med) and Red Seas. The Med brines derive from the dissolution of Messinian evaporates containing both halite (NaCl) and gypsum (CaSO4) and thus these fluids have elevated concentrations of sulfate (up to 135 mM SO42-) and hydrogen sulfide (up to 10 mM). The temperature (T) of Med brines is slightly elevated above the bottom water (by ~3ºC) and the pH is ~6.6. Red Sea brines have a geothermal source and thus have lower sulfate (10 mM), higher T (23-67 ºC), and lower pH (5.5). The GoM brines share features with both Med and Red Sea brines. GoM brines are derived from dissolution of Jurassic halite and thus contain no sulfate. Some GoM brines (e.g. GC233 and GB425, see below) are derived from the deep subsurface, so their temperature is ~10ºC or more higher than bottom water. Joye_Brine_FigA1_sonar2.jpgOther brines (e.g., GK2, see below) are derived from dissolution of salt structures near the sediment-water interface and thus have temperatures similar to the bottom water. The pH of the GoM brines is circumneutral (~7.2) and the salinity is about 150‰. The hydrogen sulfide concentration of GoM brines (<1 mM) is generally lower than that of Med brines (up to 10 mM). A common feature of GoM, Med, and Red sea brines is the presence of a 1 to 3 m thick chemocline separating the dense brine and the overlying seawater; this chemocline is a zone of elevated microbial abundance and activity.

Seafloor brines are extreme environments, meaning the resident microbial populations possess unique physiological adaptations to osmotic and sometimes thermal stress. Despite the harsh nature of the habitat, microorganisms often attain markedly high levels of diversity and productivity by exploiting the resources available in brines. Microbiological investigations of Med and Red Sea brines revealed diverse and metabolically active microbial communities that were distinct from the microbial communities in the overlying water column. In most Med brines, Bacteria were more abundant than Archaea. Med brines contained a variety of Bacteria, including representatives from the γ-, δ-, and ε- subdivisions of the Proteobacteria; Halobacteria; KB1, a new candidate division of the Euryarchaeota; MSBL-1 (Mediterranean Sea Brine Lakes group 1); and several new candidate divisions of Bacteria, including MSBL-2 through 6. Candidate division MSBL-2 is related to candidate division SB1, which was discovered along the chemocline of the Shaban Deep, Red Sea. Microorganisms related to the ANME-1 group of anaerobic methane oxidizers were found in Med brines and a diverse array of halophilic or halotolerant microorganisms were cultured from the chemocline and brine; two of these isolates were less than 92% similar to known microorganisms. In terms of microbial activity, substantial rates of sulfate reduction and methanogenesis were documented and rates of other processes (e.g. fermentation) were presumed to be substantial. The microbial diversity and assorted metabolic capabilities documented in Med and Red Sea brines illustrates that seafloor brines are a fruitful habitat for identifying new and unusual microorganisms.

Our previous work documented microbial diversity and activity in sediments influenced by brine and/or oil and gas seepage. We determined patterns of microbial abundance and activity across a 2m chemocline separating the overlying seawater and underlying brine at a mud volcano and a brine pool (Joye et al. 2009 Nature Geoscience).New Brine Figure 3_web_0.jpgGas seepage was apparent at both sites, though it was more vigorous at the mud volcano; the methane was of mixed biogenic/thermogenic origin. We documented striking differences with respect to biogeochemistry and microbial activity between the two brines. Microbial abundance in both brines was 100 times higher than the overlying seawater. The concentration of dissolved organic carbon (DOC) was elevated in the brine, suggesting a deep geosphere source. The brines contained extremely high concentrations of dissolved hydrogen (H2; Fig. 3 A, C); concentrations were higher in the brine pool (maximum=6 µM) than in the mud volcano (maximum=1µM). High H2 concentrations imply the presence of an actively fermenting community and/or limited H2 consumption.

The elevated H2 and acetate concentrations together with acetate-δ13C data implied homoacetogenesis was more important in the brine pool than in the mud volcano. Rates of sulfate reduction were 10 times higher in the brine pool than in the mud volcano while rates of methanogenesis were almost 1000 times higher in the mud volcano. No AOM was detected, possibly because of the high H2 concentrations. Data describing the identity and diversity of the microorganisms in the brines is, at this time, limited to the sulfate reducers. The surface layers of brine pool and mud volcano harbor sulfate-reducing bacterial populations that oxidize acetate and aromatic compounds (Desulfosarcinales, Desulfobacterium). The brine pool contained sulfide- and hydrogen-oxidizing epsilonproteobacteria as well, consistent with observed higher hydrogen levels and with sulfide production via sulfate reduction (Fig. 3). In contrast, the mud volcano contained fermentative deltaproteobacteria of the family Syntrophaceae that grow syntrophically with hydrogen-consuming methanogens; this is consistent with lower hydrogen concentrations and higher rates of hydrogenotrophic methanogenesis observed there (Fig. 3). 

This work generated a variety of intriguing questions, such as: What novel microorganisms inhabit these systems? Why does H2 accumulate to such high concentrations in the brine pool? Why are sulfate reduction rates lower in the mud volcano? Why are methanogenesis rates lower in the brine pool? How important are homoacetogenesis and sulfur disproportionation in these brines? We are addressing these and other questions to understand the relationship between brine chemistry/dynamics and microbial community composition and activity during this Microbial Observatory project.

We are applying a suite of microbiological, biogeochemical, isotopic, molecular, and physiological techniques to document the activity and distribution of microorganisms at brine seeps. Our overall goal is to identify and analyze the microbial populations inhabiting brine seeps characterized by different fluid flow rates and chemical regimes. 

Our work has six main objectives:

  1. Quantify the abundance, diversity, and activity of microorganisms mediating carbon and sulfur transformations in brine seep habitats;

  2. Quantify the relationship between microbial diversity and activity and environmental gradients;

  3. Identify the environmental controls on microbial activity and distributions;

  4. Identify metabolically active microorganisms;

  5. Elucidate carbon and sulfur flow in the microbial food web using stable isotope studies of biomarkers, DNA and RNA;

  6. Isolate and characterize microorganisms from brine seep habitats.

The Nuts and Bolts of the Brine Microbial Observatory Project

We are conducting research cruises and laboratory studies to examine microbial diversity and activity at seafloor brine seeps and are addressing the following hypothesis and questions at four sites over a 5-year research program (2009-2014): We hypothesize that differences in fluid flow generate variability in brine chemical composition, which strongly influences microbial community composition and activity within and between sites.

Joye_Brine_FigB2_0.jpgThis hypothesis implies that identifying and quantifying the physical and chemical constraints of seafloor brine seeps and their effects on microbial community composition and activity is essential for developing an ecosystem-level understanding of brine-impacted cold seeps, which are globally distributed marine habitats with the unique role of re-injecting fossil carbon into the living biosphere.


The experimental design exploits the different fluid flow rates and chemistry of the study sites. Fluid flow rates area characterized by modeling temperature and salinity profiles collected during each visit and monitored between cruises using in-situ sensors to document episodic events. Chemical composition, microbial abundance and community structure, and rates of microbial activity are quantified using standard methods. Metabolically active microorganisms will be identified using molecular (biomarker; RNA) and stable isotope tracer (13C SIP) methods. Standard techniques are used to isolate and characterize novel microorganisms.

The following specific research goals are being examined to address the primary hypothesis and research objectives (noted in parentheses):

  1. Spatio-temporal Variability: Document differences in microbial populations and activity between brine pool and mud volcano sites and between years within a given site (objectives 1, 2).
  2. Chemical Composition: Determine how brine chemical composition influences microbial abundance, community composition, and activity (objectives 2, 3).
  3. Functional and Phylogenetic Diversity: Quantify the microbial metabolisms and key functional genes found across habitats and evaluate how metabolic diversity relates to overall phylogenetic diversity as determined by 16S rDNA and rRNA analysis (objectives 1, 4).
  4. Regulation of Microbial Activity: Elucidate the factors—community composition, physiological constraints (e.g., limitation by nutrients or bioactive trace metals, inhibition by salt, or competition for substrates), or environmental factors—that regulate the distribution and activity of microorganisms in seafloor brines (objective 3).
  5. Flow of Energy and Carbon: Determine which chemicals serve as primary sources of metabolic energy and cell carbon to support microbial growth (objective 5).
  6. Regulation of Sulfur Isotopes: Quantify sulfur isotopic fractionation during sulfate reduction to understand the sulfur isotope composition of sulfate and reduced sulfur (objective 5).
  7. Novel Microorganisms: Isolate and characterize brine microorganisms (objective 6).

Research Cruises

Brine_Fig4a.jpgCold seeps in the northern Gulf of Mexico have been sampled extensively using submersibles and remotely operated vehicles (ROVs). This work focuses on sites spanning a depth range between 600 and 2500m; the fluid flux regimes vary from low to high (Fig. 4). The four main study sites are influenced by the seepage of halite-derived, gas-charged brines; the gas is of mixed biogenic/thermogenic origin. Oil seepage also occurs at the mud volcano sites. 1. Low fluid flux, brine pool: Site GC233 (700 m) is a quiescent brine pool surrounded by a dense community of methanotrophic mussels; nearby sediments are covered with Thiomargarita and Beggiatoa mats. Site AC601 (2300 m) is a brine pool where the surrounding sediments are inhabited by burrowing sea urchins, chemosynthetic tubeworms and clams, and sulfide-oxidizing bacteria, mainly Thioploca. The Orca Basin is a stratified basin that has anoxic brine filling the deepest depths. This site is structurally similar to some of the Med and Red Sea brine basins. 2. High fluid flux, mud volcano: Site GB425 (600 m) is an actively venting mud volcano. The macro-chemosynthetic community (e.g., mussels, clams, and tubeworms) is patchily developed, but sulfide-oxidizing bacterial mats are common on shoreline sediments. Site GK2 (1920m) is a mud volcano characterized by vigorous gas and mud discharge. The GK2 brine has a temperature similar to the bottom water, distinguishing it from GB425, where the brine temperature is elevated substantially above bottom water temperature.

Two additional sites representing “end member” brines, discovered during JASON ops in 2007, are being sampled. Site 1 (“Hot Site”) is a small mud volcano at a depth of 1000m. The central brine pool was highly active, discharging copious gas and fluidized mud. The fluid salinity was approximately 210‰ and the temperature was elevated substantially above the bottom water. Site 2 (“Red Crater”) is an ~1-km wide crater at a depth of 2200m. The edges of the crater were noted by dark-stained depressions filled with reducing brines. The central crater was filled with reddish-colored brine (salinity ~39‰). We suspect that site 1 represents the freshly formed mud volcano since the salinity approaches that expected for halite saturation. Site 2 appears to have undergone dilution with seawater over extended time given the low salinity. At these sites, we are doing CTD casts through the brine and collect bottle (or 1 brine-trapper) samples from the core brine and chemocline to evaluate rates of key microbially-mediated processes and to generate some preliminary data on microbial community composition.

We are sampling these sites during cruises using the R/V Altantis and manned submersible ALVIN in 2010 and 2013. Cruises consist of sampling each site and recovering and deploying monitoring equipment. For the four main sites, a minimum of two dives per site is required to accomplish site photo-documentation, brine sampling using the “brine trappers,” temperature-salinity profiling using a CTD, and deployment of monitoring equipment. Two sets of brine-trapper samples are being collected at three sites, while three sets of brine-trapper samples are being collected at the other site (the intensively sampled site) to provide a measure of spatial heterogeneity and obtain material for laboratory experiments.


We are using temperature-salinity profiles, temperature-salinity loggers, and time-lapse digital video to document spatial and temporal variability of fluid flow at seafloor brine seeps. Modeling of temperature and salinity profiles provides estimates of fluid flow velocities. The aim of this work is to document the variability in and magnitude of fluid flux at the different sites, not to constrain fluid flow in an absolute sense; though the latter is a worthy goal, it is well beyond the primary scope of this project, which is to understand the microbiology and biogeochemistry of these habitats. We are prepared to modify our plans to take advantage of opportunities to sample episodic or extreme discharge events at mud volcanoes. At the four main sites, the following tasks are being completed during two ROV dives (three dives at the intensively sampled site): 1. The site is being surveyed and photo-documented using a down-looking digital mosaic camera. 2. Brine samples are being collected using the brine trapper; additionally, a deep (>5m) brine sample is being collected using a gas-tight fluid sampler lowered into the brine using a winch. 3. A SeaBird SBE53-MP CTD (modified for hypersaline conditions) is being lowered into the brine using a ROV-mounted winch to characterize the thermal and salinity structure; at least 4 CTD profiles are being obtained to characterize spatial variability. 4. Monitoring equipment, i.e. a rotary time-lapse camera system and T-S sensor strings, are being deployed at positions in (sensors) or at the edge of (camera) the brines to document the T-S regime (sensors) and discharge of turbid fluid and gas bubbles (camera) over a year. Cameras and temperature-salinity loggers are being deployed at two sites per year, generating annual records for each of the four primary sampling sites during the field program.

A fluid profile from the overlying seawater into the brine is being obtained by lowering the brine trapper vertically, maintaining the upper three chambers in the overlying seawater while immersing the remainder in the brine. The seawater-brine interface is identified visually by noting the change in refractive index. The brine trapper is positioned and then sampling occurs (the chambers are opened, then closed). Using ALVIN’s dynamic positioning system, we can place the two brine trappers at comparable locations with respect to the brine-seawater interface, assuring that the contents are replicates of one other. Upon return to the surface, the brine trappers are transferred to a cold van, where they are sampled. The volume of gas in each chamber is being quantified by connecting a 20L Tedlar bag containing distilled water (and no gas phase) to the sampling port. As the pressure valve is opened, venting gas is introduced into the top of the bag while water is displaced through a one-way valve at the bottom; the volume of displaced water is recorded as a proxy for the gas volume. Three 20 mL samples of the venting gas are being collected into an evacuated serum vial for determination of concentrations of C1-C5 alkanes, carbon dioxide, hydrogen, and hydrogen sulfide (methods in the following section). The gas volume and concentration of individual components is being used to calculate in situ gas concentrations. After degassing, fluid samples from replicate chambers (the comparable depth from each trapper) is being transferred to argon-purged sterile bottles, and aliquots are dispensed for various analyses (see below).

How we will do this: Methods                         

During cruises, a suite of core experiments are carried out and environmental data is obtained at each study site. More detailed sampling and experimentation is carried out at one of the study sites each year so that detailed data is generated for each of the four main study sites over the 4-yr field program. This work focuses on the microbial communities mediating carbon and sulfur transformations in anoxic brines, the anoxic chemocline (seawater-brine transition), and the oxic overlying seawater.

1. Microbiology and molecular ecology

The goals of this component are to document patterns of microbial abundance, diversity, and associations, to characterize and compare/contrast the microbial communities over depth and between sites, and to isolate and characterize novel microorganisms. These goals are related to objectives 1, 2, 4, and 6. Detailed phylogenetic analyses cannot be completed on every sample given the large number of samples this work generates (a minimum of ~32 brine samples are being collected at each site [2 sets of replicate brine trappers, volumes combined, and 2 deep brine bottle samples], or ~130 samples total for the four main sites). For comparisons between sites, we are carrying out detailed phylogenetic characterization of the microbial community in the overlying water, in the middle of the chemocline, and in the deep brine. Cell counts and CARD-FISH enumerations are being carried out more frequently. We are using DGGE as a survey tool to identify discontinuities in microbial community structure along chemical and depth gradients to pinpoint representative samples for diversity analysis using clone libraries and pyrosequencing. Culture work is being carried out using chemocline and deep brine samples.

2. Microbial activity

Quantifying rates of microbial activity is necessary to fulfill objectives 1, 2 and 3. We are determining rates of C and S cycling in samples by assaying a core suite of activities at each site in the complete brine trapper profile samples (SR, AOM, methanogenesis from CO2 and acetate, homoacetogenesis) complemented by other assays carried out at fewer depths (acetate oxidation, fermentation, methanogenesis from methylated amines, and methanol). Rate measurements are being conducted on sub-samples from each depth of the profile collections as well as in deep brine samples (n=3 per depth plus killed controls). Activity of SR, AOM, homoacetogenesis, and methanogenesis at simulated in situ pressure is being documented at four depths (top, middle, and bottom of chemocline, and deep brine) at the intensively sampled site each year.

3. Linking phylogeny and activity

We are using three approaches to link microbial phylogeny with patterns of activity; a secondary goal is to constrain carbon flow. This work is being done at the intensively sampled site each year and addresses objectives 4 and 5.

4. Biogeochemistry

Determining the biogeochemical signature of the brine seeps helps us understand the controls on microbial distributions and activity and helps identify feedbacks between microorganisms and the geochemical environment and vice versa. Stable C and S bulk isotopic measurements combined with compound specific isotope data helps constrain C and S flow. This work is required to address objectives 2, 3 and 4.

5. Habitat characterization: imaging and instrumentation

Environmental characterization is required to address objectives 2 and 3. The PIs have access to a variety of specialized sampling and monitoring equipment, much of it developed by members of our own research team, including:

  1. Time-lapse cameras for temporal documentation of bottom conditions and communities,
  2. Salinity and Temperature loggers for recording in situ S and T over 12 months,
  3. Brine trapper for obtaining vertically stratified samples through the upper 2 to 4 m of brine.

At each site, the ROV is surveying the perimeter of each pool while recording down-looking digital images with the JASON mosaic camera. A series of small markers (floats on short lines) is being deployed to mark the edges of the pools and to facilitate repetitive sampling. Uniform scale mosaics of the pool edges documents the stability of the margin and changes in frequency and character of the habitat. A rotary time-lapse camera system is being deployed at a position where it can survey the discharge from the pool and regions of the pool edge colonized by bacterial mats and/or chemosynthetic macrofauna (e.g. Bathymodiolus mussels). These devices consist of a digital camera and light inside a tempered glass tube. The camera rotates 36º between shots to collect a 360° cylindrical panorama of the seep environment over a radius of approximately 3m. In total, this represents an area of ~28 m2 and volume of about ~43 m3 for each panorama. In the yearlong intervals between cruises, the cameras are being left in place to record time-lapse imagery for ~9 month periods (constrained by battery life). These long-term deployments are monitoring brine dynamics over annual cycles (Vardaro et al. 2006). The separate images comprising a 360° panorama is being merged into Quick Time Virtual Reality (QTVR) to offer a unique educational perspective of the brine seep environment.


Joye, S. B., V. A. Samarkin, B. N. Orcutt, I. R. MacDonald, K.-U. Hinrichs, M. Elvert, A. P. Teske, K. G. Lloyd, M. A. Lever, J. P. Montoya, and C. D. Meile, 2009. Surprising metabolic variability in seafloor brines revealed by carbon and sulfur cycling. Nature Geoscience, 2: 349-354.

Boetius, A., and S.B. Joye, 2009.  Thriving in Salt:  Life in hypersaline habitats. Science 324: 1523-1525.

Dohm, J.M., H. Miyamoto, G.G. Ori, A.G. Fairen, A.F. Davila, G. Komatsu, W.C. Mahaney, P. Williams, S.B. Joye, and 27 others, 2010. Prime candidate environments for life detection on Mars. Geological Society of America, Frontiers Volume, “Analogs for Planetary Exploration”, Special Paper 483.

Shah, S.R., S.B. Joye, J. Brandes, and A.P. McNichol, 2013. Carbon isotopic evidence for microbial control of carbon supply to Orca Basin, Gulf of Mexico. Biogeosciences, 10: 3175-3183.

Peterson, R.N., R.F. Viso, I.R. MacDonald, and S.B. Joye, 2013. On the utility of radium isotopes as tracers of seafloor hydrocarbon discharge. Marine Chemistry,

Feng, D., H.H. Roberts, S.B. Joye, and E. Heydari, 2013. Formation of low-magnesium calcite at cold seeps in an aragonite sea. Terra Nova (in press).

Crespo-Medina, M., M.W. Bowles, V.A. Samarkin, K.S. Hunter, and S.B. Joye. Microbial diversity and activity in seafloor brine lake sediments (Alaminos Canyon 601, Gulf of Mexico). The ISME Journal (submitted 18 July 2013, in review).


Vladimir Samarkin (Research Scientist)

Melitza Crespo-Medina (was Postdoctoral Researcher 2010 - 2013)

Ryan Sibert (PhD student)

Lindsey Fields (Postdoctoral Researcher)


National Science Foundation “Emerging Frontiers” program