The 2008 eruption of Chaitén volcano, Southern Chile: a tectonically controlled eruption?

The May, 2008 - present eruption at Chaitén caldera is the only example of a geophysically monitored rhyolite eruption. Geologic and seismic monitoring was conducted by the Chilean Servicio Nacional de Geología y Minería (SERNAGEOMIN) with assistance from the Volcano Disaster Assistance Program (VDAP, a joint program of USGS and OFDA). In addition, global remote sensing assets were focused on the eruption and provide extensive data on the eruptive plume and ash cloud. An initial analysis of seismic and observational monitoring and remote sensing data lead us to suggest that the Chaitén eruption was tectonically controlled, as described below. The Chaitén eruption began abruptly with Plinian ash columns (May 2-8), and then transitioned into lava dome effusion accompanied by continuous low-level ash plumes. Heights and durations of the Plinian phase of the eruption initially suggested magmatic volumes of up to as much as 1 km3, ranking this as a large VEI 4 or possibly a small VEI 5 eruption. However, reports of relatively modest thicknesses of downwind tephra indicate a smaller explosivity, probably in the moderate VEI 4 range. Extrusion of the lava dome continues at a high rate as of this time (mid-September, 2008). We estimate a lava volume of >0.3 km3 and eruption rates that have frequently exceeded 20 m3s-1, anomalously high rates for a sustained lava dome eruption. Little detailed on-site study of the proximal deposits of the eruption has been possible because of continued hazards from the eruption and austral winter weather conditions. However, several inferences about the nature of the eruption are evident. The apparent lack of historic eruptions, absence of a hydrothermal system, rapid onset of the eruption, crystal-poor rhyolite composition, lack of decompression reaction rims on amphibole crystals, and relatively high magmatic temperatures (about 860°C, as reported elsewhere in this session) all argue for rapid ascent of magma from depth. The initially high plumes and only limited column collapse argue for a relatively strong and narrow conduit and vent structure during the early explosive eruptions. However, the sustained high-rate and large-volume lava eruption (now 4 months in duration) accompanied by relatively low explosivity and low SO2 emissions argue against a purely gas-driven eruptive process. We propose a model for magmatism at Chaitén, in which the timing and compositions of eruptions are controlled by tectonism along the Liquiñe-Ofqui Fault Zone (LOFZ), a 1200 km long structure that is part of a dextral transpressional arc domain. In this model, silicic magmas are trapped and stored at deep levels of the crust (10 km?) during periods of upper crustal localized compression, and eruptions are triggered by tectonic shifts that open tear faults and promote magma transport to the surface. Consistent with seismicity along the LOFZ and subsidiary branches before and during the eruption and with new InSAR data that indicate fault-controlled syn-eruptive deformation, we suggest that such a process triggered the 2008 eruption and that re-establishment of compression, following the initial Plinian phase, has sustained an anomalously high-rate of lava production for the past four months. In turn, fluid flow along the fault-fracture network enhances seismic activity, which is still recorded. The two- way coupling between tectonics and volcanism provides a challenging conceptual framework for hazards assessment in Southern Andes.