A RHEOLOGICAL MAP OF TUNGURAHUA VOLCANO (ECUADOR): EXPLAINING THE EXPLOSIVE-EFFUSIVE TRANSITION

Tungurahua volcano has been highly active since 1999. Notably, the voluminous August 2006 eruption generated a series of explosive events, which terminated with the effusion of a lava flow. Understanding such a rapid shift in eruption style is crucial to eruption forecasting at andesitic arc volcanoes. Here, we investigate the rheological changes occurring during magma ascent to explain eruptive style.

Erupted material from the August 2006 bimodal activity is described as chemically homogeneous (with XRF analysis of the bulk rock at ~57% SiO₂). The explosive phase showed a wide range of porosities (1-60%), crystallinities (10-20% phenocrysts), and a less evolved interstitial glass composition (63-65% SiO₂). In comparison, the lava material is more crystalline (20-30% phenocrysts, high microlite content), less porous (1-5%) with an interstitial glass content of 67% SiO₂. Comparatively, the pore overpressure required to achieve fragmentation of the explosive magma was 3MPa, whereas 6-10 times more pore pressure is required to induce fragmentation of the effusive magma.

Rheological behavior of ascending magma (undergoing crystallization, volatile exsolution and chemical fractionation) is a chief determinant of eruptive style. We combined a variety of experimental techniques to map the rheological evolution of magma during ascent at Tungurahua. In the reservoir, the magma is envisaged as crystal poor and thus, has a composition similar to that of the bulk rock. We measured the non-Arrhenian temperature dependence of the viscosity of the (dry) magma in the reservoir (from remelted whole rock) as well as the increase in melt viscosity due to initial (20 vol.%) crystallisation using a concentric cylinder. The end viscosity of the erupted products was elucidated using a uniaxial press and shows an apparent viscosity 5 orders of magnitude above the pure interstitial melt and 7-8 orders above the viscosity of the magma in the reservoir. The effusive material was comparatively more viscous (and more shear-thinning) than the explosive material.

We suggest that the effusion at the terminus of the explosive phase in August 2006 resulted from the late and slower ascent of a more-viscous magma with increased crystallinity and lesser bubble load, thus with diminished stored energy to further drive the explosive eruption.