Search

Recent geophysical evidence for large-scale regional crustal inflation and localized crustal magma intrusion has made Lastarria volcano (northern Chile) the target of numerous geological, geophysical, and geochemical studies. The chemical composition of volcanic gases sampled during discrete campaigns from Lastarria volcano indicated a well-developed hydrothermal system from direct fumarole samples in A.D. 2006, 2008, and 2009, and shallow magma degassing using measurements from in situ plume sampling techniques in 2012. It is unclear if the differences in measured gas compositions and resulting interpretations were due to artifacts of the different sampling methods employed, short-term excursions from baseline due to localized changes in stress, or a systematic change in Lastarria's magmatic-hydrothermal system between 2009 and 2012. Integrated results from a two-day volcanic gas sampling and measurement campaign during the 2014 International Association of Volcanology and Chemistry of the Earth's Interior (IAVCEI) Commission on the Chemistry of Volcanic Gases (CCVG) 12th Gas Workshop are used here to compare and evaluate current gas sampling and measurement techniques, refine the existing subsurface models for Lastarria volcano, and provide new constraints on its magmatic-hydrothermal system and total degassing budget. While compositional differences among sampling methods are present, distinct compositional changes are observed, which if representative of longterm trends, indicate a change in Lastarria's overall magmatic-hydrothermal system. The composition of volcanic gases measured in 2014 contained high proportions of relatively magma- and water-soluble gases consistent with degassing of shallow magma, and in agreement with the 2012 gas composition. When compared with gas compositions measured in 2006-2009, higher relative H2O/CO2 ratios combined with lower relative CO2/St and H2O/St and stable HCl/St ratios (where St is total S [SO2 + H2S]) are observed in 2012 and 2014. These compositional changes suggest variations in the magmatic-hydrothermal system between 2009 and 2012, with possible scenarios to explain these trends including: (1) decompression-induced degassing due to magma ascent within the shallow crust; (2) crystallization-induced degassing of a stalled magma body; (3) depletion of the hydrothermal system due to heating, changes in local stress, and/or minimal precipitation; and/or (4) acidification of the hydrothermal system. These scenarios are evaluated and compared against the geophysical observations of continuous shallow inflation at ~8 km depth between 1997 and 2016, and near-surface ( < 1 km) inflation between 2000 and 2008, to further refine the existing subsurface models. Higher relative H2O/CO2 observed in 2012 and 2014 is not consistent with the depletion or acidification of a hydrothermal system, while all other observations are consistent with the four proposed models. Based on these observations, we find that scenarios 1 or 2 are the most likely to explain the geochemical and geophysical observations, and propose that targeted shallow interferometric synthetic-aperture radar (InSAR) studies could help discriminate between these two scenarios. Lastly, we use an average SO2 flux of 604 ± 296 t/d measured on 22 November 2014, along with the average gas composition and diffuse soil CO2 flux measurements, to estimate a total volatile flux from Lastarria volcano in 2014 of ~12,400 t/d, which is similar to previous estimates from 2012.

Gas-water-rock reactions taking place within volcano-hosted hydrothermal systems scrub reactive, water-soluble species (sulfur, halogens) from the magmatic gas phase, and as such play a major control on the composition of surface gas manifestations. A number of quantitative models of magmatic gas scrubbing have been proposed in the past, but no systematic comparison of model results with observations from natural systems has been carried out, to date. Here, we present the results of novel numerical simulations, in which we initialized models of hydrothermal gas-water-rock at conditions relevant to Icelandic volcanism. We focus on Iceland as an example of a "wet" volcanic region where scrubbing is widespread. Our simulations were performed (using the EQ3/6 software package) at shallow (temperature100°C. We find that this range of model gas compositions reproduces well the (H2O-CO2-STOT) compositional range of reservoir waters and surface gas emissions in Iceland. From this validation of the model in an extreme end-member environment of high scrubbing, we conclude that EQ3/6-based reaction path simulations offer a realistic representation of gas-water-rock interaction processes occurring underneath active magmatic-hydrothermal systems.

Hekla is a frequently active volcano with an infamously short pre-eruptive warning period. Our project contributes to the ongoing work on improving Hekla's monitoring and early warning systems. In 2012 we began monitoring gas release at Hekla. The dataset comprises semi-permanent near-real time measurements with a MultiGAS system, quantification of diffuse gas flux, and direct samples analysed for composition and isotopes (δ13C, δD and δ18O). In addition, we used reaction path modelling to derive information on the origin and reaction pathways of the gas emissions.Hekla's quiescent gas composition was CO2-dominated (0.8mol fraction) and the δ13C signature was consistent with published values for Icelandic magmas. The gas is poor in H2O and S compared to hydrothermal manifestations and syn-eruptive emissions from other active volcanic systems in Iceland. The total CO2 flux from Hekla central volcano (diffuse soil emissions) is at least 44Td-1, thereof 14Td-1 are sourced from a small area at the volcano's summit. There was no detectable gas flux at other craters, even though some of them had higher ground temperatures and had erupted more recently. Our measurements are consistent with a magma reservoir at depth coupled with a shallow dike beneath the summit. In the current quiescent state, the composition of the exsolved gas is substantially modified along its pathway to the surface through cooling and interaction with wall-rock and groundwater. The modification involves both significant H2O condensation and scrubbing of S-bearing species, leading to a CO2-dominated gas emitted at the summit. We conclude that a compositional shift towards more S- and H2O-rich gas compositions if measured in the future by the permanent MultiGAS station should be viewed as sign of imminent volcanic unrest on Hekla.