Dynamic Probabilistic Hazard Mapping in the Long Valley Volcanic Region CA: Integrating Vent Opening Maps and Statistical Surrogates of Physical Models of Pyroclastic Density Currents

Abstract

Ideally, probabilistic hazard assessments combine available knowledge about physical mechanisms of the hazard, data on past hazards, and any precursor information. Systematically assessing the probability of rare, yet catastrophic hazards adds a layer of difficulty due to limited observation data. Via computer models, one can exercise potentially dangerous scenarios that may not have happened in the past but are probabilistically consistent with the aleatoric nature of previous volcanic behavior in the record. Traditional Monte Carlo‐based methods to calculate such hazard probabilities suffer from two issues: they are computationally expensive, and they are static. In light of new information, newly available data, signs of unrest, and new probabilistic analysis describing uncertainty about scenarios the Monte Carlo calculation would need to be redone under the same computational constraints. Here we present an alternative approach utilizing statistical emulators that provide an efficient way to overcome the computational bottleneck of typical Monte Carlo approaches. Moreover, this approach is independent of an aleatoric scenario model and yet can be applied rapidly to any scenario model making it dynamic. We present and apply this emulator‐based approach to create multiple probabilistic hazard maps for inundation of pyroclastic density currents in the Long Valley Volcanic Region. Further, we illustrate how this approach enables an exploration of the impact of epistemic uncertainties on these probabilistic hazard forecasts. Particularly, we focus on the uncertainty of vent opening models and how that uncertainty both aleatoric and epistemic impacts the resulting probabilistic hazard maps of pyroclastic density current inundation.

Probabilistic Volcanic Hazard Assessment for Pyroclastic Density Currents From Pumice Cone Eruptions at Aluto Volcano, Ethiopia

“…This means that a wider region may potentially be affected by any given PDC, but that probability of inundation is more homogeneous across the hazard domain. This "dissipating effect" has been observed at other caldera systems (e.g., Campi Flegrei, Italy; Neri et al, 2015;Sandri et al, 2018), although the computed probability map of PDC inundation may still strongly depend on the vent-opening model used (e.g., Long Valley caldera, USA; Rutarindwa et al, 2019). By Comparison, at "central-vent" volcanoes, conditional (or even absolute temporal) PDC inundation probabilities approach 1 close to the vent [e.g., Somma-Vesuvius, Italy; (Tierz et al, 2016a,b); Soufrière Hills, Montserrat; (Dalbey, 2009)].…”

“…This presents a technical challenge, as often-used PDC simulation tools such as Titan2D (Patra et al, 2005), and VolcFlow (Kelfoun and Druitt, 2005;Kelfoun et al, 2017) are computationally expensive, allowing only a relatively limited number of runs to be performed in a reasonable time frame. Therefore, performing PVHA with such models requires the use of complex uncertainty quantification techniques such as Polynomial Chaos Quadrature (Dalbey et al, 2008;Tierz et al, 2018) or Gaussian Process emulators (Bayarri et al, 2009(Bayarri et al, , 2015Spiller et al, 2014;Wolpert et al, 2018;Rutarindwa et al, 2019). An alternative approach is to use a simple, less computationally expensive model that requires fewer assumptions to be made about the nature of the eruption.…”

“…Whatever the intensity measure, such a map can be combined with exposure data, such as population distribution, or locations of interest or importance. Additionally, if the conditional probability of PDC inundation is combined with a temporal probability of eruption, and the probability of a PDC being produced given an eruption: the non-conditional probability of PDC inundation within a given time window can be calculated (e.g., Sandri et al, 2014Sandri et al, , 2018Bevilacqua et al, 2017;Rutarindwa et al, 2019). This, combined with the exposure data, allows the calculation of the non-conditional risk (probability of asset X being lost due to PDC inundation over time period Y).…”

Major explosions and paroxysms at Stromboli (Italy): a new historical catalog and temporal models of occurrence with uncertainty quantification

“…We report the average rates in the last 10 and 25 years for comparison. www.nature.com/scientificreports/ hot avalanches 63,121,127,131 . Similarly, specific vulnerability functions should be introduced for the main hazardous actions considered 29,128 .…”

Long-Term Probabilistic Volcanic Hazard Assessment Using Open and Non-open Data: Observations and Current Issues

“…Some examples include open-access numerical codes for granular flows (TITAN2D, Patra et al, 2005; LaharFlow) 5 and uncertainty quantification techniques ranging from probabilistic graphical models (e.g., Sobradelo et al, 2014;Tonini et al, 2015;Tierz et al, 2017) to polynomial chaos methods (Dalbey et al, 2008;Tierz et al, 2018) and stochastic-process emulators (Bayarri et al, 2009;Spiller et al, 2014;Rutarindwa et al, 2019).…”