Calibration and application of silica-water triple oxygen isotope thermometry to geothermal systems in Iceland and Chile

In contrast to samples from the geothermal plant, natural amorphous silica precipitates and waters formed in active hot springs (T = 63-84 °C) in the Puchuldiza geothermal area of northern Chile gave temperature estimates from the silica-water thermometer far lower (37-46 °C) than the measured water temperatures. Active silica precipitation was found to only occur at and above the air-water interface on glass slides placed in the hot spring waters for 9 days. The calculated temperatures and visual inspection suggest that precipitation occurred along channel edges when saturation was overstepped by a factor of two. In contrast to the surficial neoformed amorphous silica, subsurface silica samples (>10 cm) have recrystallized to opal-CT and quartz within a sinter mound and these samples preserve isotope temperatures of 82 °C and 89 °C, in good agreement with the ambient temperatures of the thermal spring conduit system. The δ18O values of abiogenic, low temperature silica formed in spring water far from the thermal waters with a measured temperature of 19 °C correspond to a silica-water temperature estimate of 20 °C. All samples preserved isotope data corresponding to their expected formation temperatures and appear to be in equilibrium in the triple oxygen isotope system. A best-fit θ-T relationship for silica-water using our inorganic silica-water samples is θ=0.5305-1.82(±0.02)T(K),R2=0.998whereθa-b=lnα17Oa-blnα18Oa-b. This new equation is indistinguishable from a previous empirical fit by Sharp et al. (2016) based primarily on biogenic silica samples, suggesting that the biogenic and abiogenic samples secreted silica with the same fractionation. Our results show that triple oxygen isotope measurements are robust and can be used to estimate the temperature of formation, the isotopic composition of the formation water, and discern between equilibrium and non-equilibrium processes.