VTFS project: Development of the lava flow simulation code LavaSIM with a model for three‐dimensional convection, spreading, and solidification

Abstract

[1] The lava flow simulation code LavaSIM has been developed to give accurate predictions of volcanic disasters and to support evacuation plans. The code uses three-dimensional analysis with free surface evaluation, including boundary transport between the melt and the crust. It is therefore applicable to various types of lava flows and flow behaviors such as flood basalts, subaqueous lava flows, and levee formation. Heat transfer between the lava and the ground, air, and water, and between the melt and crust of the flow is calculated by using appropriate relations. The code has been verified by applying it to actual lava flows, and the simulation results have been compared with observations of the Izu-Osima flows of 1986. The lava flow rate, temperature, and properties were assigned values based on data from existing literature. The comparisons demonstrated the code's capability for prediction of inundated areas and maximum flow lengths. The present code is expected to be useful for real-time prediction during eruptions, in addition to assessing volcanic hazards and designing protection structures for existing installations against possible future lava flows.

Emplacing a Cooling-Limited Rhyolite Lava Flow: Similarities with Basaltic Lava Flows

“…However, despite the sophisticated numerical models that have been developed for basaltic lava flows (Crisci et al, 1986;Hidaka et al, 2005;Vicari et al, 2007;Hérault et al, 2011), a full understanding of the factors controlling the rate and extent of lava flow advance, which is essential for adequate hazard forecasting over a broad range of lava geochemistries, remains elusive due to the complexity of lava flow rheology, internal architecture, and interactions with topography. The frequency of basaltic eruptions has provided many examples for modeling low-silica lavas (Crisci et al, 1986;Hidaka et al, 2005;Vicari et al, 2007;Hérault et al, 2011), but equivalent studies of high-silica flows are relatively rare and poorly constrained, and thus reflect weaknesses in our universal understanding of lava emplacement processes. Here, we use observations of the 2011-2012 Cordón Caulle rhyolite lava flow as an unparalleled opportunity to study a high-silica flow, and we present the first modeling study to our knowledge that examines the advance of a high viscosity and crystal-poor lava.…”

Image‐based measurement of flux variation in distal regions of active lava flows

“…[3] Although the importance of flux magnitude is clear in lava flow models, and relatively long-term flux changes have been incorporated [Crisci et al, 2003;Hidaka et al, 2005], the influence of this type of rapid change is not yet known. In order to improve flow models, a greater understanding of flux variations and their causes is required.…”

Benchmarking computational fluid dynamics models of lava flow simulation for hazard assessment, forecasting, and risk management

“…Construction of cones and depositing of tephra near the vent area, flow levees along channel margins, and elongated tumuli formed by flow inflation will all affect the routing of subsequent flows (e.g., Mattox et al, 1993;Wolfe, 1988;Dietterich et al, 2012;Elissondo et al, 2016). Models should therefore be time-dependent and include syn-eruptive alteration of the topography, or they will lose accuracy as the eruption continues (e.g., Hidaka et al, 2005).…”

“…We focus on deterministic codes, but all can be applied probabilistically with Monte Carlo methods. Existing deterministic models range from onedimensional (1D; e.g., FLOWGO, Harris and Rowland, 2001) to 3D (e.g., LavaSIM, Hidaka et al, 2005;GPUSPH, Bilotta et al, 2015) and incorporate a range of physical complexity, including thermal and rheological evolution. Most models for lava flow emplacement are 2D (SCIARA, Crisci et al, 2004;MAGFLOW, Vicari et al, 2007; LavaPL, Connor et al, 2012;VOLCFLOW, Kelfoun and Vargas, 2015;COMSOL, Chen and Rempel, 2015) applying either cellular automata or a shallow-water approximation (depth-averaging) to simulate unconfined flow in plan view, but without vertical variability in parameters.…”