Structured elicitation of expert judgement in real-time eruption scenarios: An exercise for Piton de la Fournaise volcano, La Réunion island

Alessandro Tadini; Andrew J. L. Harris; Julie Morin; Andrea Bevilacqua; Aline Peltier; Willy Aspinall; Stefano Ciolli; Patrick Bachelery; Benjamin Bernard; Jonas Biren; Antonio Brum da Silveira; Valery Cayol; Oryaelle Chevrel; Diego Coppola; Hannah R. Dietterich; Amy Donovan; Olaya Dorado; Stephane Drenne; Olivier Dupere; Lucia Gurioli; Stephan Kolzenburg; Jean-Christophe Komorowski; Philippe Labazuy; Domenico Mangione; Stefano Mannini; Francois Martel-Asselin; Etienne Medard; Sophie Pailot-Bonnetat; Victoria Rafflin; Michael Ramsey; Nicole Richter; Silvia Vallejo-Vargas; Nicolas Villeneuve; Silvia Zafrilla

doi:10.30909/vol.05.01.105131

Home/Archive/USGS

USGS · 70231752

Structured elicitation of expert judgement in real-time eruption scenarios: An exercise for Piton de la Fournaise volcano, La Réunion island

View original source ↗Export metadata ↓

Abstract

Formalised elicitation of expert judgements has been used to help tackle several problematic societal issues, including volcanic crises and pandemic threats. We present an expert elicitation exercise for Piton de la Fournaise volcano, La Réunion island, held remotely in April 2021. This involved 28 experts from nine countries who considered a hypothetical effusive eruption crisis involving a new vent opening in a high-risk area. The tele-elicitation presented several challenges, but is a promising and workable option for application to future volcanic crises. Our exercise considered an “uncommon” eruptive scenario with a vent outside the present caldera and within inhabited areas, and provided uncertainty ranges for several hazard-related questions for such a scenario (e.g. probability of eruption within a defined timeframe; elapsed time until lava flow reaches a critical location, and other hazard management issues). Our exercise indicated that such a scenario would probably present very different characteristics compared to recent eruptions, and that it is fundamental to include well-prepared expert elicitations in updated civil protection evacuation plans to improve disaster response procedures.

Explore related subjects

Volcanica Article

Source-reported bounding extent: -21.423668314313243° to -20.820308271198467° latitude; 55.15960693359375° to 55.85723876953125° longitude. This indicates report coverage, not an exact sampling location. View area on OpenStreetMap.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Alessandro Tadini, Andrew J. L. Harris, Julie Morin, Andrea Bevilacqua, Aline Peltier, Willy Aspinall, Stefano Ciolli, Patrick Bachelery, Benjamin Bernard, Jonas Biren, Antonio Brum da Silveira, Valery Cayol, Oryaelle Chevrel, Diego Coppola, Hannah R. Dietterich, Amy Donovan, Olaya Dorado, Stephane Drenne, Olivier Dupere, Lucia Gurioli, Stephan Kolzenburg, Jean-Christophe Komorowski, Philippe Labazuy, Domenico Mangione, Stefano Mannini, Francois Martel-Asselin, Etienne Medard, Sophie Pailot-Bonnetat, Victoria Rafflin, Michael Ramsey, Nicole Richter, Silvia Vallejo-Vargas, Nicolas Villeneuve, Silvia Zafrilla. 2022-05-05. Structured elicitation of expert judgement in real-time eruption scenarios: An exercise for Piton de la Fournaise volcano, La Réunion island. https://doi.org/10.30909/vol.05.01.105131

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

Relating glassy rind thicknesses to ambient air temperatures at the Lost Jim flow field in the Imuruk Lake volcanic field, Alaska

The Lost Jim flow field, in the Imuruk Lake volcanic field, Alaska, extends west ~34 km from a single vent, crossing subarctic tundra and currently touches several lakes and streams. The weighted mean of five 36Cl cosmogenic exposure ages from the Lost Jim pāhoehoe flow is 7.73 ± 0.37 ka, indicating this eruption occurred substantially after the eruption of the underlying Camille flow, which was emplaced at 39.7 ± 1.3 ka. Paleoclimate records indicate the period when the Lost Jim flow field was emplaced was after deglaciation, and the climate was similar to today. We propose that the emplacement of lava in these cold subarctic conditions can lead to faster cooling of the lava surface compared to lava emplaced in warmer locations such as mid- latitude cold deserts. Glass abundance in the outermost rinds at the Lost Jim flow field was on average 74 % with 6.4 mm thick rims, compared to 60 % with 2.9 mm rims for cold mid-latitude desert samples. We interpret increased glass content as a proxy for rapid cooling likely occurring partly during winter. Glassiness values varied less across vent, margin, and mid-flow locations when compared to the mid-latitude flows suggesting the Lost Jim flow field was broadly impacted by the subarctic climate as opposed to responding to local microclimates. Our results indicate that lava glassiness may be a useful environmental indicator of cooler (in this case subarctic) conditions.

Alaska↗

USGS2026

Effects of external water on volcanic column height and collapse

Volcanic plumes are important because they spread volcanic material, can impact climate, and pose hazards to aviation. Among eruption processes, column collapse is arguably the most consequential in terms of direct impacts, as it marks the onset of ground-hugging pyroclastic density currents that pose the greatest immediate threat to life and infrastructure. This study inves- tigates how external water, such as is incorporated during subaqueous eruptions, affects the critical condition (including mass eruption rate, temperature, and gas fraction) at which column collapse occurs in the atmosphere, referred to as “the column collapse condition”. We use the 1-D plume model, Plumeria , to explore how variations in external water (0–60 wt.%), vent exit velocity (75–125 m s −1 ), and initial magma temperature (700–1100 °C) affect the column collapse condition. We find that the occurrence of column collapse is highly sensitive to the amount of external water. Small amounts of external water (⪅ 25 wt.%) suppress column collapse, whereas higher amounts of water encourage collapse. Using more than 150,000 simulations, we map out the highly non-linear shape of the column collapse condition as a function of mass eruption rates and external water contents. The Richardson number, the ratio of buoyancy to shear forces, offers a useful framework for defining the column collapse condition for eruptions involving external water. This work enhances our understanding of how external water ingestion affects volcanic plume dynamics, including height and column collapse conditions. Recent shallow submarine eruptions that produced very tall atmospheric plumes and partial column collapse (e.g. the 2021 Fukutoku-Oka-no-Ba eruption in Japan and the 2022 Hunga eruption in Tonga) demonstrate the importance of understanding how external water influences the collapse of eruption columns.

Volcanica↗

USGS2025

Case study of deep learning image segmentation for the purposes of rapid 2D petrographic analysis in volcanic rocks

Automation using deep learning methods is a useful alternative to manual methods of petrographic segmentation, but often requires user familiarity with coding and/or algorithms. We examine the Dragonfly TM program's deep learning tools for application by users with a variety of skill levels as a method for petrographic image segmentation. An image processing methodology, bimodal image stacking, was created for low-input-data, high-efficacy training of models which can then be applied to varied samples. Using backscatter electron images we show that the resulting model segmentations agree with manual segmentation total and modal crystallinity values within 5%, and calculated plagioclase crystal size distribution (CSD) values within 2σ, despite limitations in discriminating mafic phases. Model creation and training takes <24 hours, 1–3 hours of which are supervised, and the resultant model can then be applied to new uncharacterized samples in <15 minutes per image. This allows for non-experts to create and utilize deep learning models to segment images of variable brightness and texture, at low user-time cost and resulting in size and shape data which are within uncertainty of manual segmentation. While some limitations are noted (for example, sieve-textured phases may need manual correction, and different minerals with similar BSE intensity may not be resolved as separate phases), this methodology can be utilized for general application of models to wide ranges of volcanic crystalline and bubble textures, and to create a library of models for rapid petrological analysis during volcanic eruptions.

Alaska↗