2.2 CHEMISTRY

2.2a Hydrothermal Fluid Sampling - David Butterfield

The main goals for fluid sampling for this cruise were 1) re-sample vent fluids from sites sampled last year and expand the sampling to cover a broader range of the eruption area in the SE caldera and rift zone, 2) re-sample diffuse, warm vents and hot vents in the ASHES field, and 3) sample vent fluids from CASM. The vent fluid studies are intended to use chemistry to understand the connections between volcanic activity, hydrothermal heat loss, and microbial activity in hydrothermal systems. Fluid chemistry is also an important measure of habitat conditions for studies of vent fauna colonization and species distribution.

Very few studies to date have focused intensively on the variation in diffuse fluid chemistry within a region. This work may yield new insights into the variability of sub-seafloor processes in a hydrothermal system.

Hot Fluid Sampler

We used the Hot Fluid Sampler (HFS) for the second year during this cruise, after several improvements were made over the proto-type model of last year. The sampler has a lower profile in front of the ROV so that visibility and ability to sample are not severely impaired. The plumbing layout was improved to allow more filters on the instrument. This year we were configured to take 6 piston samples (3 modified with high-vacuum valves for gas sampling), 8 collapsible bag samples with optional in-line filters, and 10 additional filters without water collection. One of the bag samplers was used as a flushing line for two titanium gas-tight samplers so that fluids of known temperature could be sampled by different types of samplers for comparison. On one dive, we can collect 10 discrete water samples for fluid chemistry, 5 water samples for gas analysis, and up to 17 filters for a variety of analytical purposes.

Several types of filters were used. Membrane filters of .45 or .2 micron pore size were used to collect particulate material for: xrf chemical analysis and SEM, particulate elemental sulfur, particulate organic carbon, lipids, and fluorescent in-situ hybridization (FISH) analysis. In addition, "Sterivex" high through-put cartridge filters were used to collect particles for DNA extraction. All of the analytical work done on the filters is designed to help us understand the chemical and microbiological variation among different types of vents. The in-situ filtration capability of the fluid sampler is an important advance over all previous vent fluid sampling instrumentation in that it allows us to concentrate particulate material from large volumes of water for multiple analyses at a given vent site, and it also allows us to separate the dissolved components from the particulate components at the time of venting, thereby eliminating uncertainty about post-sampling particle formation within the water sample containers. In many instances, we collect both filtered and unfiltered water from the same site for comparison. Overall, we have an excellent set of samples for fluid chemistry, gas chemistry, particle chemistry, and microbiology.

On-board fluid analysis:

Kevin Roe was the principal fluid analyst on this cruise, and carried out analysis for pH, alkalinity, hydrogen sulfide, dissolved silica, and ammonia. Most of our chemical analysis occurs on shore. We measured refractive index of the hotter vent fluids to estimate their salinity or chlorinity. Dave Butterfield analyzed particulate elemental sulfur by colorimetry on a subset of samples.

We were fortunate to have Andy Graham on board. Andy received cuts of most of the fluid samples for methane and hydrogen analysis by gas chromatography. In many cases, vent fluid samples arrived on deck containing a gas phase and a liquid phase. When this occurred we measured the volume of the liquid and gas portions and Andy analyzed both for gas content. In this way, the fluid sampler appears to be superior to the traditional titanium major samplers because it has much better gas retention capability, and the samplers are transparent, allowing us to separate the gas portion from the liquid. We cannot rule out some gas loss through the check valves, but comparison of the gas results from different samplers will help to evaluate this.

Initial results:

HFS was deployed on 2 short aborted dives and 3 complete dives in the 1998 eruption area, plus one complete dive at ASHES. HFS proved to be quite efficient. For example, on the first HFS dive, we had only 30 minutes of bottom time at marker 33 and took 5 samples (2 water chemistry, one gas chemistry, and two filters). For the entire cruise, we collected 43 vent fluid samples for water chemistry with HFS, 13 gas piston samples, and over 50 filters. In addition, titanium gas-tight samples were collected on many dives (see Leigh Evans report) and these will be analyzed for major element chemistry.

We have not yet processed the raw data from the last two dives, but the initial results appear similar to last year. Relative to the CoAxial eruption area, where there were large changes in composition and a decay in heat output one year after the eruption, we saw little change on the seafloor in the eruption area. Diffuse venting continued in many areas. The maximum measured temperature at marker 33 is higher than last year (78 versus 55 degrees). Looking at the ratio of hydrogen sulfide to silica (the ratio is only slightly sensitive to the amount of entrained seawater during sampling) it appears that the samples from throughout the eruption area this year are just slightly lower in sulfide relative to silica. Methane appears to be slightly higher this year than last year. We cannot say anything about the changes in salinity until we do high-precision analysis on shore.

2.2b OsmoSampler and OsmoAnalyzer Operations - Thomas Chapin

Changes in the chemical composition of hydrothermal effluent after a tectonic-volcanic event have been documented (e.g., Baker et al., 1987, 1998; Butterfield and Massoth, 1994; Von Damm et al, 1995; Massoth et al., 1995; Massoth et al., in press; Wheat et al., to be submitted) and a conceptual model has been developed that theorizes the chemical evolution of venting fluids (Butterfield et al., 1997). However, the timing of these changes is uncertain. To date observations of temporal variability in the chemical composition of hydrothermal fluids has relied on repeated submersible operations and the collection of discrete samples. While this technique provides some temporal constraints, a continuous water sampler or analyzer allows one to collect more samples with limited need for costly submersible operations. Our goal for this cruise was to deploy two short-term (two weeks) and two long-term (one year) continuous sampling systems to provide temporal constraints for observing hourly to daily and weekly to monthly chemical cycles in the hydrothermal effluent. Data from these samplers and their comparison to samples collected using traditional discrete sampling techniques will allow us to determine the temporal scale of chemical change in the hydrothermal effluent as the hydrothermal system evolves and may provide constraints for understanding the physical and chemical conditions at depth and the path for fluid circulation.

Two sampling systems were deployed, OsmoSamplers and OsmoAnalyzers. OsmoSamplers are continuous water samplers that use the osmotic pressure that is created across a semi-permeable membrane by solutions of differing salinity (Theeuwes and Yum, 1976; Jannasch et al., submitted). This pressure drives water across the membrane at a speed that is dependent on the surface area of the membrane, type of membrane, salt gradient, and temperature. An excess of salt is maintained on one side of the membrane, thus only temperature affects the flow of water in the sampler. Pumps in an OsmoSampler are used to continuously draw sample through a small bore (0.8 mm id) tubing that is attached to a 40-cm-long T-handle. An additional pump was used to add acid to the sample stream in most of the OsmoSamplers. A 1.5-m-long section of tubing separates the sample intake from the pump to allow the pump to be placed in an area void of hydrothermal influence and thus minimizes temperature (pump rate) fluctuations. A temperature recorder with a resolution of 0.0018°C is attached to the T-handle to monitor the same water that is being collected by the OsmoSampler. Chemical data are obtained by retrieving the sampler, cutting the sample tubing into sections, extracting the seawater, and analyzing the seawater for chemical species of interest. Time-stamps for individual samples are determined assuming a uniform temperature at the pump that translates into a uniform rate of pumping.

OsmoAnalyzers, in contrast to OsmoSamplers, use osmotic pumps to deliver reagents into a sample stream for in situ analysis (Jannasch et al 1994). An iron OsmoAnalyzer was deployed at Marker 33 to continuously measure Fe at 15-minute intervals over the next 6-9 months.

Two long-term acid addition OsmoSamplers, deployed from the NeMO 1998 September cruise were recovered on this cruise. These samplers, one at Marker 33 and one at Milky vent, continuously collected sample for 9 months providing 163 0.5-mL samples. Analysis for major elements and trace metals will be performed later in our laboratory. Milky vent started at 9.5 C but dropped to 3.6 C by the time the OsmoSampler was recovered. There was no visible floc venting from Milky and the low temperatures at the end of the deployment indicate that the diffuse vent had died out. Marker 33, on the other hand, continued to vent hydrothermal fluids up to 80 C and was quite vigorous. The NeMO-98 long term acid addition OsmoSamplers deployments appear to have been a success and will provide one of the first long-term continuous records of the chemical signature of hydrothermal fluids.

During NeMO-99, four long-term OsmoSamplers were deployed. Three OsmoSamplers were deployed at the Marker 33, in the hottest section with temperatures up to 80 C. The OsmoSamplers consisted of one regular acid addition, one bio-OsmoSampler which has a sodium azide biocide to prevent bacterial growth, and a long term Cu-OsmoSampler which will collect samples for gas analysis. Another long-term acid addition OsmoSampler was deployed in 70 C water at Magnesia vent, just north of Milky vent. Tremendous clouds of white flocculent material were coming out of Magnesia and it looked like a snowstorm.

A 2-week record high temperature record from the Hell vent was recovered. Unfortunately the deployment of a long-term acid addition OsmoSampler and temperature probe was not successful.

References:

Baker, E. T., G. J. Massoth, and R. A. Feely. 1987. Cataclysmic hydrothermal venting on the Juan de Fuca Ridge. Nature, 329, 149-151.

Baker, E. T., G. J. Massoth, R. A. Feely, G. A. Cannon, and R. E. Thomson. 1998. The rise and fall of the CoAxial hydrothermal site, 1993-1996. J. Geophys. Res., 103, 9791-9806.

Butterfield, D.A., and G. J. Massoth. 1994. Geochemistry of north Cleft segment vent fluids: Temporal changes in chlorinity and their possible relation to recent volcanism. J. Geophys. Res., 99, 4951-4968.

Butterfield, D. A., I. R. Jonasson, G. J. Massoth, R. A. Feely, K. K. Roe, R. E. Embley, J. F. Holden, R. E. McDuff, M. D. Lilley, and J. R. Delaney. 1997. Seafloor eruptions and evolution of hydrothermal fluid chemistry. Phil. Trans. R. Soc. Lond. A, 355, 369-386.

Jannasch, H. W., K. S. Johnson and C. M. Sakamoto. 1994. Submersible, osmotically pumped analyzers for continuous determination of nitrate in situ. Anal. Chem. 66, 3352-3361.

Jannasch, H. W., C. G. Wheat, M. Kastner, and D. Stakes. 1998. Long-term in situ osmotically pumped water samplers. Deep Sea Res., submitted.

Massoth, G. J., E. T. Baker, R. A. Feely, D. A. Butterfield, R. E. Embley, J. E. Lupton, R. E. Thomson, and G. A. Cannon. 1995. Observations of manganese and iron at the CoAxial seafloor eruption site, Juan de Fuca Ridge. Geophys. Res. Lett., 22, 151-154.

Massoth, G. J., E. T. Baker, R. A. Feely, J. E. Lupton, R. W. Collier, J. F. Gendron, K. K. Roe, S. M. Maenner, and J. A. Resing. 1998. Manganese and iron in hydrothermal plumes resulting from the 1996 Gorda Ridge Event. Deep Sea Res., in press.

Theeuwes, F., and S. I. Yum. 1976. Principles of the design and operation of generic osmotic pumps for the delivery of semisolid or liquid drug formulations. Ann. Biomed. Eng., 4, 343-353.

Von Damm, K. L., S. E. Oosting, R. Kozlowski, L. G. Buttermore, D. C. Colodner, H. N. Edmonds, J. M. Edmond, and J. M. Grebmeier. 1995. Evolution of East Pacific Rise hydrothermal fluids following an oceanic eruption. Nature, 375, 47-50.

Wheat, C. G., H. W. Jannasch, F. J. Sansone, J. N. Plant, and C. L. Moyer. 1998. Hydrothermal Fluids From Loihi Seamount After the 1996 Event: A Year of Change Monitored With a Continuous Water Sampler. Earth Planet. Sci. Lett., to be submitted.

2.2c Gas Sampling - Lee Evans

The primary goal of gas sampling during the NeMO '99 expedition was direct sampling of vent fluids by way of Titanium Gastight bottles and modified pistons from the PMEL Hot Fluid Sampler (HFS). Approximately 30 useful samples were gathered and their available gas contents extracted and sealed in glass ampoules for chemical analysis. Analyses include helium isotopes, hydrogen and methane.

As with 1998's samples, the geographic coverage of sampling included the east side of the caldera along the region of the 1998 lava flow, Ashes vent field on the west side and Casm vent field to the north. Time series measurements will be possible at about 5 vent sites. The coverage of diffuse vent samples was extended southward on the east side in the direction of the vestige of the eruptive fissure.

This year's method modifications present a significant improvement as compared with those used to gather samples in 1998. At least some of 1998's collection were a bit more dilute than what is desirable. Both the plumbing scheme for titanium gastight bottles and the sample integrity of the gas piston samplers (HFS) were improved.

2.2d Studies of Dissolved Gases from Hydrothermal Vent Systems - Andy Graham

The main focus of our lab is the study of dissolved gases from hydrothermal vent systems. For this cruise I brought a gas chromatograph on board and analyzed fluid samples from the hot fluid sampler, suction sampler and a niskin bottle mounted on ROPOS. The main gases that I analyzed were dissolved hydrogen and dissolved methane. These gases are important in the hydrothermal vent community because certain microbes can oxidize these gases and use them as an energy source. Over 50 samples were analyzed ranging from 300 ^oC Inferno vent to 4 ^oC Magnesia vent. From an initial glance the data vent such as Marker 33 and Virgin Mound still appear to contain high concentrations of both hydrogen and methane. Further analysis and a comparison to last year's data will occur.