Integrated geochemical modelling of magmatic degassing and hydrothermal interaction: a case study from Kawah Ijen volcano, Indonesia
Abstract
Monitoring active volcanoes requires an understanding of magmatic degassing in relation to magma depth, temperature, composition, style of degassing (open vs closed) and interactions with hydrothermal systems. This study combines results of subsurface degassing (interpreted from melt inclusions) with measurements of fumarole gases and acid spring waters from Kawah Ijen volcano, Indonesia. Kawah Ijen is a stratovolcano with a growing rhyolite dome on the shore of a hyperacidic crater lake. The dome is emitting sulfur-rich gases from high temperature fumaroles (350-450°C). Matrix glass and melt inclusion compositions (including H2O, CO2, S, Cl and F) were measured for basaltic, dacitic and rhyolitic magmas. The behavior of the volatile species (Dvap-melt) during ascent, degassing and crystallization were modeled for an open system (including vapor fluxing) assuming Rayleigh fractionation, and for closed system processes assuming batch degassing and crystallization. The variable H2O-CO2 contents of the melt inclusions suggest that open system vapor fluxing (XH2Ovapor = 0.25-0.95 for basalt; 0.9-0.95 for dacite) is the dominant degassing style. The modeled S Dvap-melt values for basalt remain low (2-10) as the melt ascends (P= 400 to 100 MPa), then increase sharply to 200 at pressures <100 MPa. As the melt evolves from basalt to dacite, S Dvap-melt values are high (100-300) and independent of pressure. Evolution from dacite to rhyolite is characterized by a constant Dvap-melt value of 35. Chlorine behavior is strongly affected by crystallization of Cl-rich apatite in the basaltic magma. In dacite and rhyolite, Cl is mostly dissolved in the melt. The Dvap-melt values range from 7-9 as basalt evolves to dacite and reach 5 for dacite to rhyolite (low pressure degassing). Fluorine contents are highly variable due to crystallization of F-apatite, especially in the more evolved rocks. This precludes meaningful modeling of F-release to the vapor. The best-fit modeled gas compositions (mass ratio) are: CO2/H2O = 0.13-0.27, CO2/S(total) = 2.9-5.7, H2O/S(total) = 21-22 and CO2/HCl = 8-16; fumarole gas compositions are: CO2/H2O = 0.25-0.3, CO2/S(total) = 4-5, H2O/S(total) = 12-20 and CO2/HCl = 230-420. The latter represent some of the most S- and CO2-rich and Cl-poor compositions recorded for arc volcanoes. Partly, the vapor data are interpreted to reflect high S solubility in the magma due to high fO2, Ca and Fe, and the low Cl content a result of apatite crystallization. In part, they are interpreted to reflect condensation of the magmatic vapor upon ascent. S and C are concentrated in the residual vapor and Cl lost to the liquid; acid spring waters are enriched in Cl (SO42-/Cl = 3). Interaction of the gas with dome rocks causes pyritization and precipitation of native S, leading to a disproportionately low flux of SO2 (150 tonnes/day). This study illustrates the importance of integrating melt inclusion data with surface gas compositions in evaluating the effects of magmatic and hydrothermal processes on surface degassing patterns at active volcanoes.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2010
- Bibcode:
- 2010AGUFM.V53B2250V
- Keywords:
-
- 8410 VOLCANOLOGY / Geochemical modeling;
- 8419 VOLCANOLOGY / Volcano monitoring;
- 8430 VOLCANOLOGY / Volcanic gases