New chemical and original isotopic data on waters from El Tatio geothermal field, northern Chile

The El Tatio geothermal field is located at an height of 4200-4300 m on the Cordillera de los Andes (Altiplano). Geysers, hot pools and mudpots in the geothermal field and local meteoric waters were sampled in April 2002 and analyzed for major and trace elements, delta(2)H, delta(18)O and H-3 of water, delta(34)S and delta(18)O of dissolved sulfate, delta(13)C of dissolved total carbonate, and Sr-87/Sr-86 ratio of aqueous strontium. There are two different types of thermal springs throughout the field, that are chloride-rich water and sulfate-rich water. The chemical composition of chloride springs is controlled by magma degassing and by water-rock interaction processes. Sulfate springs are fed by shallow meteoric water heated by ascending gases. In keeping with the geodynamic setting and nature of the reservoir rocks, chloride water is rich in As, B, Cs, Li; on the other hand, sulfate water is enriched only in B relative to local meteoric water. Alternatively to a merely meteoric model, chloride waters can be interpreted as admixtures of meteoric and magmatic (circa andesitic) water, which moderately exchanges oxygen isotopes with rocks at a chemical Na/K temperature of about 270 degrees C in the main reservoir, and then undergoes loss of vapor (and eventually mixing with shallow water) and related isotopic effects during ascent to the surface. These chloride waters do not present tritium and can be classified as submodern (pre-1952). A chloride content of 5,400 mg/l is estimated in the main reservoir, for which delta(2)H and delta(18)O values, respectively of -78 parts per thousand and -6.9 parts per thousand, are calculated applying the multistage-steam separation isotopic effects between liquid and vapor. From these data, the meteoric recharge (Cl approximate to 0 mg/l) of the main reservoir should approach a composition of 107 parts per thousand in delta(2)H and -14.6 parts per thousand in delta(18)O, when a magmatic water of delta(2)H = -20 parts per thousand, delta(18)O = +10 parts per thousand and Cl = 17,500 mg/l is assumed. The Sr-87/Sr-86 ratios of the hot springs are quite uniform (0.70876 to 0.70896), with values within the range observed for dacites of the Andean central volcanic zone. A water delta(18)O-Sr-87/Sr-86 model was developed for the main geothermal reservoir, by which a meteoric-magmatic composition of the fluids is not excluded. The uniform delta(34)S (SO42-) values of +1.4 to +2.6 parts per thousand, in the chloride waters agree with a major deep-seated source for sulfur, possibly via hydrolysis in the geothermal reservoir of sulfur dioxide provided by magma degassing, followed by isotopic exchange between sulfate and sulfide in the main reservoir. This interpretation is supported by the largely negative delta(34)S (SO42-) value in steam-heated water sulfate (-9.8 parts per thousand) and mass-balance calculation, which exclude leaching at depth of igneous iron-sulfides with delta(34)S near zero per mill. All the delta(13)C values of total carbonate in the chloride waters are negative, with variable values from -9.2 to -20.1 parts per thousand, pointing to an important proportion of biogenic carbon in the fluids. The interpretation of these data is problematic, and a number of alternative explanations are reported in the text.