Abstract

Mineral inputs to Yellowstone Lake, Wyoming, come from a variety of sources, including hydrothermal vents, ground water, rain water, flux from sediments, and direct runoff. One third of Yellowstone Lake is directly influenced by hydrothermal activity (hot water vents and fumaroles). Geothermally heated water percolating through the chamber is highly enriched in carbonate, silicate, chloride, and methane, with some locations additionally rich in iron and sulfide. Vent waters in West Thumb typically contained sub-micromolar concentrations of Fe (iron), while those in Mary Bay and off Stevenson Island contain up to 1 mM. The water column concentrations of dissolved iron range from 250 to 450 nM (nanomolar) in Mary Bay, but were very low in the waters of Southeast Arm, West Thumb, and off Stevenson Island. Pore water and vent water chemistry provided evidence for lake water dilution of vents below the sediment-water interface. Significant fracturing of source water conduits was indicated by extreme differences in pore water profiles from cores less than 5 m apart in geothermally vigorous West Thumb. Some samples approached theoretical reservoir composition for geothermally active areas of Mary Bay and West Thumb showing chloride concentrations reaching several millimolar (mM), and in the case of Mary Bay, extrapolate to the geothermal end member (~20 mM) at a depth of only 2-3 meters. These steep concentration gradients support diffusive chloride fluxes across the sediment-water interface 3 orders of magnitude higher than those in non-venting depositional areas.

Introduction

Yellowstone Lake, WY is located in the caldera of the largest volcanic eruptions known, which occurred 1.2M and 650,000 years ago at a mid-continental hot spot, rather than the more widespread tectonic spreading centers. The Yellowstone hot spot has interacted with the North American plate for millions of years causing widespread outpourings of basalt. Some of the basaltic melt, or magma, produced by the hot spot accumulates near the base of the plate, where its heat melts the rocks from the Earth's lower crust. As a result the underlying structure is composed primarily of granite overlain by volcanic silica as opposed to freshly upheaved basalts. Geothermally heated water percolating through the relic chamber is highly enriched in carbonate, silicate, chloride and methane: some locations are also enriched with iron, manganese and sulfide. The Yellowstone National Park is well known for its steaming geysers, shimmering thermal pools and bubbling painted mudpots. Some of the greatest characteristics that are not visible are the hydrothermal vents submerged under Yellowstone Lake, hydrothermal activity in the form of springs and fumaroles were described by Remsen et al. (1990), and Marocchi et al. (2001).

The magma chamber encompasses the northern part of Yellowstone Lake, while the Yellowstone River inflow and the southern half of the lake (South and Southeast arms) are outside the caldera. Previous work has shown active hydrothermal venting (geothermal hot springs and fumaroles) in several areas of the lake, which strongly influences the chemical composition of the lakewater (Cuhel 1998, 1999; Klump 1988). This is also observed in deep-sea hydrothermal vents, where vigorous plumes mix with deep water (Butterfield et al. 1997; Cowen et al. 1986), but the large receiving volume defies budget closure, which is one of the goals of past work in Yellowstone Lake (Aguilar, et al. 1999).

Previous investigations of thermal waters from the Norris-Mammoth corridor have used different approaches to identifying sources of hydrothermal fluids. These have included the use of natural isotope tracers (e.g., H, He, Li), elemental abundances (e.g., S, Cl, Na, Ca) and also dissolved species present in these hydrothermal fluids (Fournier 1989; Palmer and Sturchio 1990; Kharaka et al. 1991; Bullen and Kharaka 1992; Fournier et al. 1992; Kharaka et al. 1992, Rye and Truesdell 1992; Sturicho et al. 1992; Lewis et al. 1997). Based on all these studies we can compare recent results to those performed several years ago in order to have a better understanding of the changing environment in the Yellowstone Lake area and other areas in the caldera.

The interactions of the geothermal systems with biology have an important role in the understanding the processes on the origins of early life. The high temperature systems may be relevant to understanding extreme environments on Earth as well as other planets and moons in our Solar System.

STUDY AREA

Sampling sites on Yellowstone Lake.

Yellowstone Lake is located in the southeast section of Yellowstone National Park, in an area with constant tectonic activity. The lake comprises an area of 341 km² and it is the largest high altitude lake in North America. The northwestern area of the lake lies inside the caldera, where the southern area as well as South and Southeast arms are located outside the caldera (Figure 1, Map of Yellowstone Lake showing selected sampling areas, West Thumb, Mary Bay, Stevenson Island, Southeast Arm, Yellowstone River inlet and outlet. The rim of the caldera is depicted by the dotted line. Core collection sites are in solid circles as follows: 01 for Mary Bay core, 03, 06 from West Thumb and 07 from Stevenson Island. This map was reproduced with the author’s permission, Marocchi et al. 2001). Several areas have been sampled through the years but all the collections mentioned in this paper were from 1998. There are areas with evident geothermal activity, such as Mary Bay, Sedge Bay, Steamboat Point, Stevenson Island, and West Thumb. All these areas have been sampled frequently as well as others such as the Yellowstone river inlet (outside the caldera, Southeast arm) and outlet (inside the caldera).

METHODS

Use of a Remotely Operated Vehicle.

Figure 2

The use of a Remotely Operated Vehicle (ROV) is critical for general survey and sampling hydrothermal vent systems in Yellowstone Lake (Fig. 2 ROV, Remotely Operated Vehicle from Eastern Oceanics used to collect vent and bottom water). The ROV designer and operator Dave Lovalvo of Eastern Oceanics is a former pilot of DSRV Alvin (Deep Sea Research Vessel) and ROV Jason (Remotely Operated Vehicle) who has produced a practical array of modular instruments for water and solid phase sampling, as well as cameras for still pictures and video (Buchholz et al. 1995; Klump et al. 1992). The areas of interest are hard to sample by conventional means. Visual

Figure 3

observations of shimmering waters are always important while exploring the bottom of the lake. When looking for evidence of vents on the surface waters, we rely on vigorous bubbling that is visible from a distance on a calm day (Figure 3, bubble field on surface waters of Mary Bay, on a calm day they can be seen from the distance. The bubbles are used to find new vent activity in different areas of the lake).

Field methods

Vent samples were collected with the Remotely Operated Vehicle, on board the R/V Cutthroat, with an articulated arm, outfitted with a thermistor probe in the end to measure the temperature of the water as it was collected. Water was collected into 2 L polycarbonate syringes, samples were then retrieved and put into smaller all plastic syringes through a 3-way valve. Samples were then returned in a cooler to the laboratory for analysis and preservation.

Figure 4

Cores were collected from the R/V Cutthroat with a 3-inch Benthos gravity corer with cellulose acetate butyrate liners (Figure 4, core from West Thumb inside a core liner, notice the darker sediment water-interface. This core is ready on the extruder to be transferred into the squeezer liner). Sediment was then transported to the laboratory and transferred with a hydraulic extruder to the Jahnke squeezer (Jahnke 1988) to subsequently obtain pore water (Figure 5 below, porewater squeezer used to obtain porewater by applying pressure vertically and the water tends to be forced horizontally “guided” by the porex inside the sediment at the end of the filter.

Figure 5

The picture shows how the squeezer is put together, showig the depth intervals to obtain porewater from different depths in the core). Porex inserts (porous polyethylene rod to “guide” the water through it while being pushed out by the action of the piston), were acid washed and rinsed through many changes of E-Pure water (18 meg ohm/cm resistance) to zero residual chloride. The last rinses with E-pure water were done in a Coy anaerobic chamber (90 % N₂, 10 % H₂) with water devoid of oxygen. All parts contacting the sample were acid washed and those inserted were maintained anaerobically until the instant of use (in sealed serum vials). In-line 25 mm filters (0.2 µm pore size) were Ion Chromatography-approved ultraclean commercial units (IC Gelman Acrodiscs), and all-polypropylene syringes received the sample. Components for reduced sulfur analysis were prepared in an anaerobic chamber, with dilution blanks, standards, and reagents in serum vials. Samples for trace metals were acidified with trace metal certified nitric acid and stored in acid washed polypropylene tubes. The samples for routine chemical analysis were stored at 4^oC in polypropylene tubes. Core processing (sectioning, squeezing) was accomplished in a protected part of the United States Park Service garage.

Chemistry analyses

In the laboratory samples were filtered through 0.2 µm filters (Supor, Nuclepore) and water was aliquotted for the different analyses.Dissolved mineral compounds were measured in the field laboratory by several methods: flow injection analysis (FIA: silicate, SiO₂), ion chromatography (IC: chloride, Cl^-, sulfate, SO₄⁼), spectroscopy (ammonium, NH₄⁺), according to standard methods (APHA 1992). Reduced and total iron was also determined in the field by the ferrozine spectrophotometric method of Stookey (1970), with (total iron) and without (reduced iron, FeII) reductant extraction. Total carbon dioxide, SCO₂, was analyzed by the teflon-membrane flow injection method of Hall and Aller (1992). Reduced sulfur compounds (hydrogen sulfide, H₂S, thiosulfate, S₂O₃⁼, sulfite, SO₃⁼) were quantified by a scaled-up modification of the micro-bore high performance liquid chromatographic (HPLC) method of Vairavamurthy and Mopper (1990), using dithio-bis-nitropyridine (DTNP) derivatization. The analytical equipment was transported to Yellowstone National Park, where all labile specie were analyzed on site, within one day of collection and analytical preparation.

Pore water flux was calculated from pore water concentration profiles and concentration gradients at the sediment-water interface were used to calculate fluxes via Fick’s first law of diffusion (Berner 1980): J = D_S ^.f^. dC/dz, where J is the flux of the different components; D_S is the molecular diffusivity corrected for porosity (Li and Gregory 1974); f is the porosity at the sediment-water intreface; and dC/dz is the slope of the concentration gradient.

RESULTS

Porewater

Since almost a third of Yellowstone Lake is directly influenced by hydrothermal activity, it is important to measure chemical components that can provide a proxy for geothermal activity in Yellowstone Lake. Chloride is an important indicator of geothermal activity, and the Yellowstone river inlet provides a low chloride concentration (<7 µM). The subsurface deep reservoir containing fluids that feed the thermal basins in Yellowstone National Park are thought to be about 20 to 21 mM chloride (Truesdell et al. 1977, Fournier 1989).

Porewater profiles in all graphs will depict distinct sites in Yellowstone Lake, all cores were collected during the 1998 season. Core 01-MB (open squares) was taken from a Mary Bay vent field, and smelled of hydrogen sulfide as we brought it onto the vessel. This core was close to one that melted the plastic core liner (temperature >135 ^oC) moments before. Core 03-WT (open circles) was collected near the West Thumb geyser basin. Core 06-WT (closed circles) was collected in the West Thumb deep basin. Core 07-SI (closed squares) was collected from the deep canyon East of Stevenson Island (Figure 1).

Chloride is a conservative and non-reactive ion that is used as a geothermal tracer. Chloride concentration in Mary Bay reached 10mM, the highest concentration measured in pore water (Figure 6, porewater profile depicting chloride concentration (µM, micromolar) with depth in 4 different cores from Mary Bay, West Thumb, and Stevenson Island). A concentration of about 5 mM chloride was also found in a core from West Thumb, all the other sites measured showed a concentration lower than 1 mM.

Diatoms (algae) require silica to produce frustules (skeleton made of silica). When these organisms die they settle to the bottom of the lake, during and after settling they undergo dissolution. Evidence of this process is found in the porewater profiles from the sediments in different areas of the lake. Silica is an element that is non-conservative and reactive. Silicate reflects the diagenetic/dissolution control, where decomposition takes place without geothermal influence, although there are some examples of cores with this influence. Mary Bay and West Thumb 03 cores had the highest concentrations, about 25 mM, the other core had a concentration of 1 mM (Figure 7, porewater profile depicting silicate concentration (µM, micromolar) with depth in 4 different cores from Mary Bay, West Thumb, and Stevenson Island). Silicate shows a higher concentration than expected from a diagenesis generated curve, showing the influence of vent activity in the area. The values for the Southeast arm reach a concentration of 750 µM, similar to the concentration of core 06 from West Thumb and 07 from Stevenson Island.

Hydrogen sulfide is a compound that we refer to as the “smell of success” since it is a great marker for reducing conditions in sediments as well as vent water. It is a readily distinguishable reduced component that will be present in an area where there is no oxygen present. It is the second to the last (methane) most difficult to produce. Hydrogen sulfide concentration was highest, 550 µM, in the Mary Bay core (Figure 8, porewater profile depicting hydrogen sulfide concentration (µM, micromolar) with depth in 4 different cores from Mary Bay, West Thumb, and Stevenson Island). The concentration in the other cores was less than 10 µM, which is significantly lower than the active areas. Sulfate reduction from bacteria is an important component in the production of this reduced compound. Hydrogen sulfide has been found consistently in Mary Bay.

Bacterial sulfate reduction is a process of organic matter decomposition, where sulfate is used as energy source by bacteria, by which sulfate is reduced to hydrogen sulfide. As mentioned previously, this process occurs in the absence of oxygen. Sulfate was highest, 200 µM, at Stevenson Island, the concentrations in the other cores were less than 80 µM (Figure 9, porewater profile depicting sulfate concentration (µM, micromolar) with depth in 4 different cores from Mary Bay, West Thumb, and Stevenson Island). West Thumb core 03 showed a very shallow gradient compared to the gradient from core 01.

Reduced iron concentrations were highest in the Stevenson Island core as well as one of the West Thumb (06) from the deep basin with a concentration of 37 µM (Figure 10, porewater profile depicting reduced iron concentration (µM, micromolar) with depth in 4 different cores from Mary Bay, West Thumb, and Stevenson Island). Iron laminations are found extensively in the West Thumb area (Figure 8), typically vent water lacks reduced iron in the effluent, but some areas in the sediment show evidence of iron oxides.

Ammonium is formed by the decomposition of organic nitrogen compounds, in areas where oxygen is not present, there is no oxidation to nitrite and nitrate and therefore it tends to accumulate. Diffusion is due to molecular processes only, and the concentrations show a diagenetic process (Figure 11, porewater profile depicting ammonium concentration (µM, micromolar) with depth in 4 different cores from Mary Bay, West Thumb, and Stevenson Island). Mary Bay shows a very steep gradient with a large ammonium reservoir in the sediment, a similar pattern is also observed in West Thumb 06 core.

As mentioned earlier carbon dioxide is another indicator of geothermal activity. High concentrations were measured in the Mary Bay 01 and West Thumb 03 cores (Figure 12, porewater profile depicting total carbon dioxide concentration (µM, micromolar) with depth in 4 different cores from Mary Bay, West Thumb, and Stevenson Island), both with evidence for active geothermal influence, based on the chloride concentration.

Ventwater

Vents are very heterogeneous and have a wide range of temperatures, ranging from 20 ^oC to 112 ^oC, pH from 4 to 8.6, and chemistry that varies with location. Chemical differences from vents in different areas have allowed us to group the different characteristics into domains. Vents waters from West Thumb and Mary Bay showed enrichment in chloride and silicate, although vents in West Thumb were highly variable (Table 1, Selected chemistry of Yellowstone National Park vents, Yellowstone River inlet and outlet, and water column values). Reduced iron was present in vents from Stevenson Island and Mary Bay, where the reduced species can remain in the water for at least 24 hours (data not shown).

Table 1. Selected chemistry of YNP vents, Yellowstone River, and water column.

Location	pH	Cl	Fe
		µM	µM
WT vents	5.5 - 8.6	50 - 1147	<0.18 - 0.54
MB vents	4.9 - 5.9	144 - 169	0.23 - 9.3
SI vents	5.0 - 6.2	136 - 148	1.7 - 8.1

WT water column	7.3	154	<0.18
MB water column	6.75	179	0.25 - 0.45
SI water column	7.4	141	<0.18

YR-inlet	7.05	7	0.5
YR-outlet	7.29	126	0.197

Lake water

Lake water collected by submersible in a deep vent area (Stevenson Island) showed chemical enrichment in several minerals (chloride, silicate, sulfate, sodium, etc.), when compared to surface water collected at the Southeast Arm inlet and the Yellowstone River outlet (Table 1). When lake water values were compared to vents of the different areas, it becomes evident that for example, Mary Bay has water that still reflects the hydrothermal composition of the vents.

There were distinct differences in the composition of hydrothermal vent fluids from different parts of Yellowstone Lake. For example, vents from the West Thumb area were rich in chloride but poor in sulfur compounds, as compared to vents from Stevenson Island which were rich in sulfur but poor in chloride. In contrast, chimney structures from these vents record times that the vent fluids must have been different in composition because they contain phases that could not have formed from the vent fluids that were emanating from these sites; chimney structures from Stevenson Island contain sulfur crystals as well.

Flux from the sediment into the overlying water can be calculated from the pore water chemistry from Mary Bay, West Thumb and Stevenson Island. Table 2 shows the calculated flux from chloride as the geothermal activity tracer, and silica, as the dissolution/diagenetic control in pore water. Chloride flux was highest (2 orders of magnitude) from the Mary Bay and West Thumb hot cores, other cores and areas such as Stevenson Island as well as Southeast arm (which is outside the caldera) do not provide chloride to the receiving lake water. Silica does not show such a dramatic difference, but the same cores have high standing silica concentrations throughout the cores.

Table 2. Porewater concentrations and flux from cores obtained in Mary Bay, West Thumb, Stevenson Island and Southeast Arm. Showing values for chloride, a “geothermal tracer”, and silica, a “dissolution/diagenetic control” parameter.

Station	Chloride “geothermal tracer”			Silica “dissolution/diagenetic control”
Porewater chemistry:	[conc]@ z=∞ mmol/L	Grad µM cm^-1	FluxMol m^-2 y^-1	[conc]@ z=∞mmol/L	Grad µM cm^-1	FluxMol m^{-2 .} y^-1
Mary Bay “hot” core	12.5 –20.0	450	2.41	2000	200	0.80
W. Thumb “hot” core	7.5	360	1.93	2050	224	0.90
West Thumb	0.185	5.5	0.017	1200	165	0.47
Stevenson Island	0.180	-0.6	-0.002	720	311	0.89
Southeast Arm	0.172	0.34	0.001	900	140	0.40

DISCUSSION

Mineral inputs to Yellowstone Lake come from a variety of sources, namely hydrothermal vents, groundwater, rainwater, flux from sediments and direct runoff (including tributaries). Approximately one third of Yellowstone Lake is directly influenced by hydrothermal activity (hot water vents and fumaroles. Surveys of lake water, vent water and sediment porewater gradients established zones of direct and subsurface inputs of geochemically altered fluids. Vent water intrusion into the surrounding sediments is evident in some of the profiles. In some instances, chloride approaches reservoir concentrations (15 mM) and the silicate concentration at depth seems greater than diagenesis-generated flux. Porewater and vent water chemistry provides evidence for lake water dilution of vents below the sediment-water interface.

Reduced sulfur compounds are important components of the vent waters in Mary Bay and Stevenson Island, while in West Thumb these compounds were usually undetectable. The vent fluids exhibit a highly variable concentration of dissolved minerals in different areas of the lake as well as on different years of sampling. As shown in the solid phase from West Thumb (Figure 13, solid phase sample collected from West Thumb. Note the laminations on the surface of alternating manganese and iron oxides), where highly laminated iron/manganese oxide crusts are found, in areas which typically do not contain sulfide, methane or other reduced compounds.

Strong evidence for vent fluid seepage was found in the hot core pore water measurements of chloride (10 mM), total CO2 (to 11 mM), and silicate (280 µM), both highly enriched in deep reservoir fluids. Some areas of the lake contain high sulfide concentrations (500 µM) and iron (50 µM). Because inorganic nitrogen (ammonium) is virtually absent from the water column and vent fluids, diagenetic production of ammonium from organic matter may provide more growth-promoting habitats in surrounding sediments tan in aqueous environments.

One of the factors that seemed to influence the vent activity throughout the lake was the amount of water in the lake. There seemed to be a correlation between high activity in the vents when water levels were low, and a low activity when water levels were high. This is one of the factors the will benefit from long-term studies of the different vent areas in the lake.

Based on the input data from the Yellowstone River in and the output of the river in the northern part of the lake, it is clear that there is a significant hydrothermal influence in the lake map (Figure 14, Yellowstone lake map showing different concentrations of selected compounds and the changes incurred from the source of the water coming into the lake outside the caldera region, to the Yellowstone River out). Chloride is virtually absent in the inlet waters, and when the concentration is higher than lake water, there is strong evidence for an external source of ion. During three years of piezometer studies to measure the groundwater inputs to the lake, we concluded that the flow is not sufficient to explain the concentration differences. Again, another piece of evidence that points to a geothermal influence in the concentration of key components. There are also sources and sinks of other elements, but just to mention a few we can see that this is a very dynamic system where different sources of chemicals are found and where microbiology is an important component of this dynamic geoecosystem.

ACKNOWLEDGEMENTS:

U.S. National Park Service at the Yellowstone National Park, Wyoming, John Varley, John Lounsbury, and personnel at the Aquatic Resource Center for their help in providing space for the laboratory and use of the R/V Cutthroat, including Dan Mahoney, Jim Ruzycki, Brian Ertel; Rick Fey, and Harlan Kredit (Fishing Bridge Visitors Center). The group is also thankful for having access to the Utah dormitory facility, it has been a great place to go after long hours in the laboratory. We have had the opportunity to bring different groups of undergraduates each year, and we could not do all the work without their help, we thank them for their effort: Austin Johnson, Janine Herring, Erin Breckel, Jeremy Claisse, Michelle McElvaine.

This work was supported by National Science Foundation through the Environmental Geochemistry and Biogeochemistry award EAR9708501, and Research Experience for Undergraduates award OCE 9423908 and OCE 9732316. University of Wisconsin-Milwaukee Great Lakes WATER Institute Contribution number ###.

References

Aguilar, C., Cuhel, R., and D. Lovalvo. 1999. Pore Water, Vent Water and Ground Water Inputs to Geothermally-Active Yellowstone Lake, Wyoming. ASLO Meeting, Santa Fe, February 1999.

APHA. 1992. Standard methods for the examination of water and wastewater: Washington, D.C., 18th ed. American Public Health Association, 905 p.

Berner, R.A. 1980. Early Diagenesis: A Theoretical Approach. Princeton University Press, New Jersey.

Buchholz, L. A., Anderson, P. D., Cuhel, R. L., Klump, J. V., Kostka, J. E., Paddock, R. W., Remsen, C. C., Maki, J. S., and Lovalvo, D. 1995. Employment of ROV techniques and Scuba in Yellowstone Lake, in Harper, D. E. J., ed., Diving for Science: Nahant, MA, American Academy of Underwater Science, p. 1-7.

Bullen T.D. and Kharaka Y.K. 1992. Isotopic composition of Sr, Nd, and Li in thermal waters from the Norris-Mammoth corridor, Yellowstone National Park and surrounding region, ., in Kharaka Y.K and Maest A.S. (eds.) Proceedings of the 7th International Symposium on Water-Rock Interaction; Volume 2, Moderate and High Temperature Environments, 897-901.

Butterfield, D.A., Jonasson, I.R., Massoth, G.J., Feely, R.A., Roe, K.K., Embley, R.E., Holden, J.F., McDuff, R.E., Lilley, M.D., and J.R. Delaney. 1997. Seafloor eruptions and evolution of hydrothermal fluid chemistry. Philosophical Transactions of the Royal Society London A. 355: 369-386.

Cowen, J. P., Massoth, G. J., and Baker, E. T. 1986. Bacterial scavenging of Mn and Fe in a mid- to far-field hydrothermal particle plume: Nature, v. 322, p. 169-171.

Cuhel R.L., Klump J.V., Remsen C.C., Aguilar C., Maki J., Lovalvo D. 1998. Geothermally driven chemosynthesis in Yellowstone Lake. Yellowstone National Park 125th Anniversary Symposium, May 11-23 1998, Bozeman, MT, Yellowstone Science 6(2): 32.

Cuhel RL, Maki JS, Kostka JE, Klump JV, Remsen CC. 1999. Chemosynthesis, photosynthesis, and heterotrophic production in geothermally-active Yellowstone Lake, Wyoming. Limnology and. Oceanography (in preparation).

Fournier R.O. 1989. Geochemistry and dynamics of the Yellowstone National Park hydrothermal system. Annual Review of Earth and Planetary Sciences, 17, 13-53.

Fournier R.O., Thompson J.M. and Hutchinson R.A. 1992. The geochemistry of hot spring waters at Norris Geyser Basin, Yellowstone National Park, U.S.A., in Kharaka Y.K and Maest A.S. (eds.) Proceedings of the 7th International Symposium on Water-Rock Interaction; Volume 2, Moderate and High Temperature Environments, 1289-1292.

Hall, P. O. J., and Aller, R. C. 1992. Rapid, small-volume, flow injection analysis for SCO₂ and NH₄⁺ in marine and freshwaters: Limnology and Oceanography, v. 37, p. 1113-1119.

Jahnke, R.A. 1988. A simple, reliable, and inexpensive pore-water sampler. Limnology and Oceanography 33: 483-487.

Kharaka Y.K., Mariner R.H., Bullen T.D., Kennedy B.M., Sturchio N.C. 1991. Geochemical investigations of hydraulic connections between the Corwin Springs Known Geothermal Resources Area and adjacent parts of Yellowstone National Park. in Sorey M.S. (ed.) U.S.G.S. Water Resources Inv. Report 91-4052, F1-F38.

Kharaka Y.K., Mariner R.H. and Evans W.C. 1992. Compositions of gasses from the Norris-Mammoth corridor, Yellowstone National Park, U.S.A.: Evidence for magmatic source near Mammoth hot springs, in Kharaka Y.K and Maest A.S. (eds.) Proceedings of the 7th International Symposium on Water-Rock Interaction; Volume 2, Moderate and High Temperature Environments, 1303-1307.

Klump, J. V., Remsen, C. C., and Kaster, J. L. 1988. The presence and potential impact of geothermal activity on the chemistry and biology of Yellowstone Lake, Wyoming, in De Luca, M. P. and Babb, I., ed., Global Venting, Midwater, and Benthic Ecological Processes. National Undersea Research Program Research Reports, p. 81-98.

Klump, J. V., Paddock, R., and Lovalvo, D. 1992. A 16-loop, ROV-controlled, in situ water sampler: Journal of Great Lakes Research, v. 18, p. 309-316.

Lewis A.J., Palmer M.R., Sturchio N.C., Kemp A.J. 1997. The rare earth element geochemistry of acid-sulphate-chloride geothermal systems from Yellowstone National Park, Wyoming, U.S.A., Geochimica et Cosmochimica Acta, 61, 695-706.

Li, Y-H. and S. Gregory. 1974. Diffusion of ions in sea water and in deep-sea sediments. Geochimica et Cosmochimica Acta 38: 703-714.

Marocchi, S., Remsen, C.C. and J.V. Klump. 2001. Yellowstone Lake: Join the Expedition! Hammockswing Publishing, Whitefish Bay, Wisconsin.

Palmer M.R. and Sturchio N.C. 1990. The boron isotope systematics of the Yellowstone National Park (Wyoming) hydrothermal system: A reconnaissance. Geochimica et Cosmochimica Acta, 54, 2811-2815.

Remsen, C. C., Klump, J. V., Kaster, J. L., Paddock, R., Anderson, P., and Maki, J. S. 1990. Hydrothermal springs and gas fumaroles in Yellowstone Lake, Yellowstone National Park, Wyoming: National Geographic Research, v. 6, p. 509-515.

Rye R.O. and Truesdell A.H. 1992. The question of recharge to the geysers and hot springs of Yellowstone National Park, Wyoming, U.S.A., in Kharaka Y.K and Maest A.S. (eds.) Proceedings of the 7th International Symposium on Water-Rock Interaction; Volume 2, Moderate and High Temperature Environments, 1345-1348.

Sturchio N.C., Murrell M.T., Pierce K.L., Sorey M.L. 1992. Yellowstone travertines: U-series ages and isotope ratios (C, O, Sr, U), in Kharaka Y.K and Maest A.S. (eds.) Proceedings of the 7th International Symposium on Water-Rock Interaction, Volume 2, Moderate and High Temperature Environments, 1427-1430.

Stookey, L. L. 1970 Ferrozine-a new spectrophotometric reagent for iron: Analytical Chemistry, v. 42, p. 779-781.

Truesdell, A.H., Nathenson, M., and Rye, R.O. 1977. The effects of subsurface boiling and dilution on the isotopic compositions of Yellowstone thermal waters: Journal of Geophysical Research, 82, 3694-3704.

Vairavamurthy, A., and Mopper, K. 1990. Determination of sulfite and thiosulfate in aqueous samples including anoxic seawater by liquid chromatography after derivatization with 2, 2'-dithiobis (5-nitropyridine): Environmental Science and Technology, v. 24, p. 333-337.