Skip to content

Does incorporation of irregular bomb shapes significantly influence the outcome of ballistic hazard models?

Amilea Sork 1, Ben Kennedy1, Rebecca Fitzgerald2, Kae Tsunematsu3, Marina Bisson4, Leighton Watson1, Mathieu Sellier1, Jacopo Taddeucci5, Daniele Andronico6, Piergiorgio Scarlato5, Elisabetta Del Bello5, Tullio Ricci5, Dona Banerjee1, Paul Kreit1,7

  • Affiliations: 1University of Canterbury, Christchurch, New Zealand; 2GNS Science, Lower Hutt, New Zealand; 3Yamagata University, Yamagata, Japan; 4Istituto Nazionale di Geofisica e Vulcanologia, Sezione Pisa, Pisa, Italy; 5Istituto Nazionale di Geofisica e Vulcanologia, Sezione Roma 1, Roma, Italy; 6Istituto Nazionale di Geofisica e Vulcanologia, Osservatorio Etneo, Catania, Italy; 7University of Konstanz, Konstanz, Germany

  • Presentation type: Talk

  • Presentation time: Tuesday 09:15 - 09:30, Room R280

  • Programme No: 3.15.7

  • Theme 3 > Session 15

Abstract

Volcanic ballistic projectiles (VBPs) are a near-vent volcanic hazard that can be frequently fatal. A critical component of VBP hazard assessment is the estimation of impact locations; this is often accomplished via physics-based computational models such as Ballista. Drag is one of the most significant physical controls on a VBP's impact location and is in turn controlled mainly by the VBP's shape, represented as a dimensionless drag coefficient CD. However, little has been quantified about the drag behaviour of molten VBPs (or bombs), which, especially compared to solid blocks, have irregular and often fluctuating shapes. Therefore, we present a case study comparing a modelled distribution which incorporates a CD for irregularly-shaped bombs to a real-world scenario. Firstly, we quantified the irregularity of bomb shapes. By examining high-speed video of Strombolian eruptions, we define a framework of shape definitions for bombs and a shape-size relationship. Secondly, we measured CD for realistic bomb shapes. Using 3D-printed models of our bomb shape end-members, we quantify CD ranges for each end-member in a wind tunnel analogue experiment. Thirdly and finally, we incorporate these results into the Ballista model using eruption parameters matching Stromboli's July 2019 paroxysm. We partition the modelled particles' CDs based on the size-shape distribution and incorporate our experimental drag coefficient ranges for each shape. We then compare the simulated impact distribution to the mapped distribution from Stromboli's July 2019 paroxysm and explore the influence of the bomb shape compared to other factors on the VBPs' impact locations.