Global mapping of eruptive styles to magma viscosity and ascent rate
Olivier Bernard1, Sri Budhi Utami2, Fidel Costa1
Affiliations: 1Institut de Physique du Globe de Paris, Paris, France 2School of the Environment, The University of Queensland, Brisbane, Australia
Presentation type: Poster
Presentation time: Tuesday 16:30 - 18:30, Room Poster Hall
Poster Board Number: 96
Programme No: 3.16.23
Abstract
Magma ascent rates (MAR) and viscosity are primary controls of eruption dynamics, to accurately determine either parameter from eruption deposits has proven to be challenging, and almost impossible during eruptions. Combining these two parameters for a large dataset of past eruptions with hindsight can be beneficial to determine the physical processes behind eruption styles, which is key for hazard assessment. Here, we compile magma ascent and decompression rates values from past studies, that we combine with new calculations of melt and magma viscosity for a large number of eruptions with different styles. Our curated dataset includes 222 MAR from 93 eruptions at 46 volcanic systems with magma composition ranging from basalt to rhyolite. These values were obtained using 13 different petrological or geophysical methods and represent 10 eruption styles. We find that for high-viscosity endmembers, eruption styles are bimodal, with plinian eruptions dominating at high MAR and extrusive eruptions dominating at low MAR. The MAR limit between the two domains of Plinian and extrusive decreases with higher viscosity which highlights a strain-rate control of fragmentation. Towards the low-viscosity endmember, controls are less clear, where high MAR still tends to produce intense explosivity (fountaining, plinian eruptions), whereas low and intermediate MAR show high variations of eruption styles from effusive to fountaining. Our dataset demonstrate that no eruption can occur from low MAR and low viscosity endmembers, underlining a condition linked to gas decoupling from ascending magma that prevents it from reaching the surface.