Chemical and stable isotopic models for boundary creek warm springs, southwestern Yellowstone National Park, Wyoming

Abstract

Abstract Thermal springs of the Boundary Creek hydrothermal system in the southwestern part of Yellowstone Park outside the caldera boundary vary in chemical and isotopic composition, and temperature. The diversity may be accounted for by a combination of processes including boiling of a deep thermal water, mixing of the deep thermal water with cool meteoric water and/or with condensed steam or steam-heated meteoric water, and chemical reactions with surrounding rocks. Dissolved-silica, Na + , K + and Ca 2+ contents of the thermal springs could result from a thermal fluid with a temperature of 200 ± 20°C. Chloride-enthalpy and silica-enthalpy mixing models suggest mixing of 230°C, 220 mg/l Cl − thermal water with cool, low-Cl − components. A 350 to 390°C component with Cl − ≥ 300 mg/l is possibly present in thermal springs inside the caldera but is not required to fit observed spring chemical and isotopic compositions. Irreversible mass transfer models in which a low-temperature water reacts with volcanic glass as it percolates downward and warms, can account for observed pH and dissolved-silica, K + , Na + , Ca 2+ and Mg 2+ concentrations, but produces insufficient Cl − or F − for measured concentrations in the warm springs. The ratio of a Na / a H , and Cl − are best accounted for in mixing models. The water-rock interaction model fits compositions of acid-sulfate waters observed at Summit Lake and of low-Cl − waters involved in mixing. The cold waters collected from southwestern Yellowstone Park have δD values ranging from −118 to −145 per mil and δ 18 O values of −15.9 to −19.4 per mil. Two samples from nearby Island Park have δD values of −112 and −114 per mil and δ 18 O values of −15.1 and −15.3 per mil. All samples of thermal water plot significantly to the right of the meteoric water line. The low Cl − and variable δD values of the thermal waters indicate isotopic compositions are derived by extensive dilution with cold meteoric water and by steam separation on ascent to the surface. Many of the hot springs with higher δD values may contain in addition a significant amount of high-D, low-Cl − , acid-sulfate or steam-heated meteoric water. Mixing models, Cl − content and isotopic compositions of thermal springs suggest that 30% or less of a deep thermal component is present. For example, the highest-temperature springs from Three Rivers, Silver Scarf and Upper Boundary Creek thermal areas contain up to 70% cool meteoric water and 30% hot water components, springs at Summit Lake and Middle Boundary Creek spring 57 are acid-sulfate or steam-heated meteoric water; springs 27 and 48 from Middle Boundary Creek and 49 from Mountain Ash contain in excess of 50% acid-sulfate water; and Three Rivers spring 46 and Phillips could result from mixing hot water with 55% cool meteoric water followed by mixing of acid-sulfate water. Extensive dilution by cool meteoric water increases the uncertainties in quantity and nature of the deep meteoric, thermal component.