[ Stable Isotopes ] [ ICP-MS ] [ Home ]

Temperature of precipitation for the aragonite-water system was determined with the assumption that our analyzed solid and water samples were in equilibrium. Based on this assumption, the average calculated precipitation temperature is ~57°C. Our water samples had measured temperatures much lower than this (around 30°C). Therefore, by fixing the temperature at 30°C, we solved instead for the d¹⁸O_Water equilibrium value with respect to analyzed d¹⁸O_Aragonite values. The calculated value for d¹⁸O_Water at equilibrium is -0.7 SMOW, considerably lower than the analyzed values of our samples (Table 1). This low calculated d¹⁸O_Water value suggests that our waters are not precipitating solids in the conditions under which they were collected. Factors affecting the analyzed d¹⁸O_Water value may be reduced meteoric water contribution or greater evaporative effects, both plausible for samples collected during the dry season.

In addition, temperature of precipitation was calculated for the calcite-HCO₃^- system. Assuming d¹³C_Calcite and d¹³C_Water were in equilibrium, a precipitation temperature of ~155°C was obtained. Such a high temperature seems unlikely and suggests that our waters are not in equilibrium. By fixing the temperature at 53°C (a more reasonable precipitation temperature calculated for the calcite-water system), a d¹³C_Water equilibrium value of -1.4 PDB was obtained with respect to d¹³C_Calcite. Factors which contribute to our much more negative d¹³C_Water numbers (Table 1) may include oxidation of hydrocarbons and decay of plant materials.

For the water samples we analyzed d³⁴S, D/H, d¹⁸O, and d¹³C (Table 2). The d³⁴S, d¹⁸O, and D/H values all suggest a common reservoir as the source of the water at sites 3, 4, and 8. The isotopic signature of site 5 indicated either different processes or a separate water supply. However, all samples originated as meteoric water, which has undergone geothermal heating based upon their D/H and d¹⁸O values (Figure 1). In addition to geothermal heating, some samples have undergone modification at/near the surface due to microbial activity (sulphate reduction and carbon fixation) as well as degassing (Figure 2).

The shift in dD and d¹⁸O values away from regional meteoric water values can also be explained by mixing of hydrothermal fluid with meteoric water. The geochemical evidence supports the mixing of hydrothermal waters with meteoric water. Hydrothermal waters of the region fall within two main categories, Franciscan and Great Valley (Donnelly-Nolan et al., 1993). Selected geochemical data for these two types as well as regional meteoric water can be seen in Table 2, along with data from the sampled sites. The data from the sites suggest that the spring water is a mixture of hydrothermal fluid and meteoric water. Based upon the high chloride content of the spring waters and their B/Cl ratios, the source of the hydrothermal fluids is the Great Valley Sequence.

Comparison of the samples amongst themselves allowed the differentiation of sample 5-3 from samples 3-2 and 8-1. All of the trace element data for the three waters sampled can be seen in Tables 3 and 4. Sample 5-3 contained higher concentrations of B, Na, Al, Sr, Ba, La, and Br, but lower concentrations of Li, Mg, Ni, Rb, Cs, and Si. Samples 3-2 and 8-1 appear more closely related geochemically.

Previous work in the area (See Figure 3 and Table 5) includes samples collected near sites 3-2 and 8-1 (Enderlin, 2001). The data correlate well between sites, especially 3-2 and 1550KA, suggesting that the water geochemistry is consistent from year to year. However, both samples were collected in September, so seasonal variations (possibly related to banding seen within the travertines) would not have been evident in the waters collected.

Geothermometry using the ICP-MS results and the methods of Henley et al. (1984) allowed for another comparison amongst the waters. The results of the geothermometry can be seen in Table 6. The Na/K geothermometers give the highest temperatures for the waters, whereas the quartz geothermometers give the lowest. The wide range in temperatures indicates that silica is possibly being lost due to quartz deposition in the subsurface. This would explain the lower temperatures since it is unlikely that a Na/K phase would be deposited. Another possiblity is the interaction and mixing of meteoric water with the hydrothermal waters coming from depth.

The composition and mineralogy of this travertine are interpreted to be controlled by both evaporative processes and the presence of serpentinite. ICP-MS data collected on solid samples support this conclusion in several ways.

The presence of evaporative salts in some samples is suggested by high concentrations of alkali elements and the correlation between K⁺ and Na⁺ (Figure 4). The data points are colored according to site location, which elucidates the pattern that was expected. The lowest values of K⁺ and Na⁺ correspond to site 7-1, which contained the least detrital serpentine and hosted active carbonate precipitation at the time samples were collected. The highest values correspond to sites 1-2, and 2-1, both of which contain an abundance of detrital serpentine, but were not observed to be actively precipitating carbonate. The most obvious explanation is that these elements are precipitating as evaporative salts, which is further supported by scanning electron microscopy observations of halite and silvite crystals in several of the samples.

If evaporation is concentrating Na⁺ and K⁺, it should be influencing the travertine chemistry in other ways as well. Microprobe data indicate the presence of sepiolite, which is a mineral commonly formed in evaporative environments where it typically follows the precipitation of carbonates. Sepiolite also seems to be more abundant in the presence of larger amounts of detrital serpentine. ICP-MS data support both of these ideas. The behavior of Mg²⁺ and Si correlate well (Figure 5), which can be attributed to the precipitation of sepiolite (Mg₄(OH)2Si₆O₁₅^. 6H₂O). When these data points are colored according to site location (Figure 6), the distribution of sepiolite is very similar to that of the evaporative salts discussed previously. Sepiolite is more abundant in samples that also contain abundant detrital serpentine (Mg₃Si₂O₅(OH)₄). Serpentinite in the subsurface may be important in water-rock interactions that contribute Mg²⁺ to the fluids. As these fluids are concentrated by evaporation, the Mg²⁺ is available for precipitation in the sepiolite. Note that the low Mg²⁺ and Si values in 2-2 are likely due to the extensive weathering of that site, which may have removed both sepiolite and serpentine.

Microprobe and XRD data demonstrate that aragonite is the dominant carbonate mineral in the travertine, but that both high magnesium calcite and dolomite are also present. This clearly indicates that Mg²⁺ ions are not only precipitating within sepiolite. Given the prominent role of evaporation in the salt and clay chemistry described above, this process is a reasonable explanation for calcite and dolomite precipitation as well. Figure 7 plots the Mg²⁺/Ca²⁺ ratio against the weight percent Sr²⁺ of each carbonate subsample drilled from travertine samples. The samples cluster into at least two groups, which correspond to the inclusion of calcite within the drilled carbonate bands. The lower slope of the subsamples from 7-1 suggests that the end-member high Mg²⁺ calcite from 7-1 has significantly less Mg²⁺ than the high Mg²⁺ calcite or dolomite of the samples from other sites. A lower Mg²⁺ concentration in 7-1 subsamples may suggest either that the travertine was not subjected to extensive evaporation during the precipitation of these bands, or that the source waters reaching this area were lower in Mg²⁺ concentration. A combination of these two is also possible.

References

Donnelly-Nolan, J. M., Burns, M.G., Goff, F.E., Peters, E. K., and Thompson, J.M., 1993, The Geysers-Clear Lake area, California: Thermal water, mineralization, volcanism, and geothermal potential: Soc. Econ. Geol. Guidebook Series, v. 16.

Enderlin, D., 2001, Explanation of map of known springs and seeps: McLaughlin Mine, from McLaughlin Mine Closure Plan, September 30, 2001.

Henley, R. W., Truesdell, A. H., and Barton, P.G., Jr., 1984, Fluid-Mineral Equilibria in Hydrothermal System, Reviews in Economic Geology, v. 1, p. 267.

[ Field Images ] [ Mineralogy and Magnetics ]
[ Petrography and Microbiology ]

[ GEL 281 Home Page ]
schiffman@geology.ucdavis.edu