Chlorine degassing during the lava dome-building eruption of Mount St. Helens, 2004-2005

Abstract

Remote measurements of volcanic gases from the Mount St. Helens lava dome were carried out using OpenPath Fourier-Transform Infrared spectroscopy on August 31, 2005. Measurements were performed at a site ~1 km from the lava dome, which was used as a source of IR radiation. On average, during the period of measurement, the volcanic gas contained 99 mol percent H 2 O, 0.78 percent CO 2 , 0.095 percent HCl, 0.085 percent SO 2 , 0.027 percent HF, 4.8×10 -4 percent CO, and 2.5×10 -4 percent COS close to the active vent. The fluxes of these species, constrained by synchronous measurements of SO 2 flux, were 7,200 t/d H 2 O, 140 t/d CO 2 , 22 t/d SO 2 , 14 t/d HCl, 2.0 t/d HF, 54 kg/d CO, and 59 kg/d COS, ±20 percent. Observations of H 2 O/Cl in the vapor and melt are compared to models of closed- and open-system degassing and to models where a closed system dominates to depths as shallow as ~1 km, and gases are then allowed to escape through a permeable bubble network. Although several features are consistent with this model--for example, (1) H 2 O/Cl in the gases emitted from stagnant parts of the lava dome, (2) the concentration of Cl in the matrix glass of erupted dacite, and (3) the glass H 2 O/Cl--the gases emitted from the active part of the lava dome have much higher H 2 O/ Cl than expected. These higher H 2 O/Cl levels result from a combination of two factors (1) the addition of substantial amounts of ground water or glacier-derived H 2 O to the gases at shallow depths, such that only ~10 mol percent of the measured H 2 O is magmatic, and (or) (2) some Cl present as alkali chloride (NaCl and KCl) in the gas phase. The mean molar Cl/S is similar to gases measured at other silicic subductionzone volcanoes during effusive activity; this may be due to the influence of Cl in the vapor on S solubility in the melt, which produces a solubility maximum for S at vapor Cl/S ~1.