PREEVENTS Eruption Forecasting Project

Explosive volcanic eruptions pose an increasing hazard to a growing, globally connected population. Alaska has ~54 historically active volcanoes with multiple explosive eruptions per year that threaten local populations, infrastructure, and air traffic. Accurate forecasting of volcanic eruptions is challenging due to the complex and dynamic nature of volcanoes. Recent studies, however, have shown that combining multidisciplinary observations of past unrest can yield accurate forecasts of eruptions from well-monitored volcanoes. Our team at UAF was awarded a National Science Foundation Prediction of and Resilience against Extreme Events (PREEVENTS) award from 2020-2026 to re-analyze and combine multidisciplinary observations of past volcanic unrest for eight Alaska volcanoes to develop a suite of eruption forecast models. The ultimate goal of this work was to provide eruption forecast models to the Alaska Volcano Observatory (AVO) for operational use to improve AVO’s eruption forecasting capabilities and mitigate eruption hazards. We also developed publicly available classroom modules and a distance-based continuing education course that was shared with teachers from communities near active volcanoes to help prepare the populations most directly impacted by volcanic eruptions.

This project capitalized on the wealth of AVO data; the multidisciplinary expertise of volcano observatory scientists; and the analytical power provided by a team of graduate students, postdoctoral fellows and researchers. Research and data analysis was conducted within the framework of four disciplinary themes: (1) geodesy,  (2) seismology & infrasound, (3) volcanic gas geochemistry & remote sensing, and (4) petrology, geochemistry & physical volcanology, and focused on eight recently active Alaska volcanoes (Figure 1). The graduate student cohort worked with UAF faculty, staff, postdoctoral fellows and collaborators to first analyze data within each discipline and then synthesize the data across disciplines. The multidisciplinary data were then used by two postdoctoral fellows to develop eruption forecasting models. 

Seismic data comprises one of AVO’s monitoring pillars with over 220 stations deployed across the arc, providing rich data primed for detailed study. Most eruptive phenomena carry a distinct seismic signature suitable for eruption forecasting and modeling. Infrasound, or low frequency sound below 20 Hz, is produced when volcanic eruptions perturb the atmosphere and create sound waves. Volcano infrasound provides information on eruption processes within the shallow conduit or eruption column. Similar to seismic, different eruption styles have unique, distinctive infrasonic signatures. We analyzed seismic and infrasound data with a set of comprehensive, automated tools utilizing modern techniques to provide data products and metrics before, during, and after eruptions from our target volcanoes. 

Volcano geodesy investigates ground deformation around volcanoes, which may indicate pressure changes within volcanic systems, magma migration between them, or formation of new conduits or intrusions. In this project rwe eanalyzed existing GPS data and Satellite Aperature Radar (SAR) acquisitions (including new Sentinel 1A/B), and automated the (near) real-time inversion of deformation data for volcanic sources. 

Geological observations and geochemical analyses of historical eruption products can help constrain conceptual, geochemical, and physical eruption models and characterize eruptive behavior to inform both short and long-term eruption forecasts. In this project we used petrographic observations, mineral phase thermobarometry, and mineral diffusion/dissolution profiles to provide insights into magma properties and storage conditions, the timing of magma recharge and ascent prior to eruption, and the likely eruption trigger.

We then combind and synthesized our volcano-specific data reanalysis to develop a range of simple to complex models designed to help understand these volcanic systems.  Simple conceptual models were developed by combining seismoacoustic, geodetic, remotesensing and  geochemical observations and models (Figure 2). Additionally, one complex multiparameter physics-based and four probabilistic eruption forecast models were developed, the later which incorporated geologically-derived eruption histories, geophysical and geochemical monitoring data, and expert elicitation to forecast the possible eruption outcomes and threats.

Figure 2: Conceptual model of the shallow conduit system at Pavlof volcano and its influence on eruption explosivity from Tan et al. (2025). (A) Slow magma ascent leads to effusive-dominated eruptions out of a flank vent, while (B) rapid magma ascent leads to explosive eruptions out of the summit crater.

Physical models use physical equations and realistic properties to help constrain subsurface volcano conditions that may lead to eruption. In this project we use a 2D finite element method model to investigate steam flow through fractured volcanic domes to better understand the processes driving changes in thermal anomalies at volcanic domes.

Statistical methods have been applied successfully to volcanic systems to develop probabilistic eruption forecast models. Here we apply anomaly detection methods on time series seismic, acoustic, radiative power and volcanic gas datasets and combine these results with expert elicitation identified priority monitoring metrics  for four of our target volcanoes to develop probabilistic eruption forecasts.

The following Alaska volcanoes targeted in this study encompass a range of eruptive behaviors and signs of unrest. In particular we selected four frequently erupting, open-system volcanoes (from west to east: Cleveland, Shishaldin, Pavlof and Veniaminof) and four intermittently erupting, closed-system volcanoes (from west to east: Okmok, Bogoslof, Augustine and Redoubt). Through data reanalysis and multidisciplinary synthesis we aimed to advance our understanding of these dynamic systems and identify commonalities in unrest signatures.

Kyle Anderson, U.S. Geological Survey - California Volcano Observatory

Israel Brewster, University of Alaska Fairbanks Geophysical Institute - Alaska Volcano Observatory

Cheryl Cameron, Alaska Division of Geological & Geophysical Surveys  - Alaska Volcano Observatory

Laura Clor, U.S. Geological Survey - Volcano Emissions Project

Michelle Coombs, U.S. Geological Survey - Alaska Volcano Observatory

Scott Crass,  Alaska Division of Geological & Geophysical Surveys - Alaska Volcano Observatory

Hannah Dietterich, U.S. Geological Survey - Alaska Volcano Observatory

Matt Haney, U.S. Geological Survey - Alaska Volcano Observatory

Alicia Hotovec-Ellis, U.S. Geological Survey - California Volcano Observatory

Peter Kelly, U.S. Geological Survey - Volcano Emissions Project

Christoph Kern, U.S. Geological Survey - Volcano Emissions Project

Matt Loewen, U.S. Geological Survey - Alaska Volcano Observatory

Jake Lowenstern, U.S. Geological Survey - Volcano Disaster Assistance Project

Steve McNutt, University of South Florida

Warner Marzocchi, University of Naples Federico II

Tina Neal, U.S. Geological Survey - Alaska Volcano Observatory

Jeremy Pesicek, U.S. Geological Survey - Volcano Disaster Assistance Project

John Power, U.S. Geological Survey - Alaska Volcano Observatory

Janet Schaefer, Alaska Division of Geological & Geophysical Surveys - Alaska Volcano Observatory

Lori Schoening, University of Alaska Fairbanks Geophysical Institute

Patsy Shaha, Former Resident of Perryville, Alaska

The Native Village of Perryville

Bryan Witten, University of Alaska Fairbanks Geophysical Institute

Kristi Wallace, U.S. Geological Survey - Alaska Volcano Observatory

Angarita, M., Grapenthin, R., Plank, S., Meyer, F.J. and Dietterich, H., 2022. Quantifying large‐scale surface change using SAR amplitude images: crater morphology changes during the 2019–2020 Shishaldin Volcano eruption. Journal of Geophysical Research: Solid Earth, 127(8), p.e2022JB024344. https://doi.org/10.1029/2023GC011341.

Angarita, M., R. Grapenthin, S. Henderson, M. Christoffersen, K. Anderson (2024), Versatile Modeling Of Deformation (VMOD) Inversion Framework: Application to 20 years of observations at Westdahl Volcano and Fisher Caldera, Alaska, US, Geochem., Geophys., Geosys. (G3), vol 25(4), e2023GC011341, https://doi:10.1029/2023GC011341.

Angarita, M., Wasser, V., Grapenthin, R., Lerner, A., Lopez, T., Parameswaran, R., Lyons, J. (2026), Erupted volume during the 2014–2023 Semisopochnoi unrest and eruption constrained by multi-platform InSAR and volcanic gas observations, G-Cubed, doi.10.1029/2026GC013083.

Angarita, M., Grapenthin, R., Tan, D., Saunders-Shultz, P., Girona, T., Parameswaran, R.M., Lopez, T.M., Izbekov, P.E., Fee, D., Burgos, V., Moshrefzadeh, J.A., Shreve, T., Larsen, J., Multidisciplinary perspectives on Shishaldin Volcano: An open-system deforming volcano, (Accepted), Bulletin of Volcanology, doi.10.31223/X5DR13.

Crozier, J., Karlstrom, L., Montgomery-Brown, E., Angarita, M., Cayol, V., Bato, M.G., Wang, T.A., Grapenthin, R., Shreve, T., Anderson, K. and Astort, A., 2023. Understanding the drivers of volcano deformation through geodetic model verification and validation. Bulletin of Volcanology, 85(12), p.74. https://doi.org/10.1007/s00445-023-01687-4.

Fee, D., Tan, D., Lyons, J., Sciotto, M., Cannata, A., Hotovec-Ellis, A., Girona, T., Wech, A., Roman, D., Haney, M. and De_Angelis, S., 2025. A generalized deep learning model to detect and classify volcano seismicity. Volcanica, 8(1).

Lopez, T., Lerner, A., Franco, B., Loewen, M., Kelly, P., Kern, C., Clor, L., Werner, C., Clarisse, L., Rasmussen, D., Wasser, V., Kushner, S., Saunders-Shultz, P., Brewster, I., Theys, N., Izbekov, I., (in revision), Insights into magmatic processes at Shishaldin Volcano based on satellite and airborne volcanic gas measurements and magmatic degassing models, Bulletin of Volcanology.

Lyons, J.J., Tan, D., Angarita, M., Loewen, M.W., Lopez, T., Grapenthin, R., Hotovec-Ellis, A.J., Fee, D. and Haney, M.M., 2025. Identifying precursors and tracking pulses of magma ascent in multidisciplinary data during the 2018–2023 phreatomagmatic eruption at Semisopochnoi Island, Alaska. Journal of Volcanology and Geothermal Research, 463, p.108329. https://doi.org/10.1016/j.jvolgeores.2025.108329.

Moshrefzadeh, J., Izbekov, P., Loewen, M., Larsen, J. and Regan, S., 2023. Dating individual zones in phenocrysts from the 2016–2017 eruption of Bogoslof volcano provides constraints on timescales of magmatic processes. Journal of Volcanology and Geothermal Research, 435, p.107741. https://doi.org/10.1016/j.jvolgeores.2022.107741.

 Saunders-Shultz, P., Lopez, T., Dietterich, H. and Girona, T. (2024), Automatic identification and quantification of volcanic hotspots in Alaska using HotLINK: The hotspot learning and identification network. Frontiers in Earth Science, 12, p.1345104, https://doi.org/10.3389/feart.2024.1345104.

Tan, D., Fee, D., Hotovec-Ellis, A.J., Pesicek, J.D., Haney, M.M., Power, J.A. and Girona, T., 2023. Volcanic earthquake catalog enhancement using integrated detection, matched-filtering, and relocation tools. Frontiers in Earth Science, 11, p.1158442. https://doi.org/10.3389/feart.2023.1158442.

Tan, D., Fee, D., Witsil, A., Girona, T., Haney, M., Wech, A., Waythomas, C. and Lopez, T., 2024. Detection and characterization of seismic and acoustic signals at Pavlof Volcano, Alaska, using deep learning. Journal of Geophysical Research: Solid Earth, 129(6), p.e2024JB029194. https://doi.org/10.1029/2024JB029194.

Tan, D., Fee, D., Izbekov, P., Lopez, T., Girona, T., Burgos, V., McNutt, S., Haney, M., Wasser, V., Larsen, J., Grapenthin, R., Angarita, M., Saunders-Shultz, P., Shreve, T., Moshrefzadeh, J., (2025) Vent-specific unrest at Pavlof Volcano, Alaska: Insights from Multidisciplinary data, Journal of Volcanology and Geothermal Research, 468, https://doi.org/10.1016/j.golgeores.2025.108465.

Wasser, V.K., Izbekov, P.E., Lopez, T.M., Graham, N.A. and Moshrefzadeh, J.A., 2025. Using dissolution surfaces in crystal clots to quantify pressure changes preceding the 2006 eruption of Augustine Volcano, Alaska. Geology, 53(8), pp.689-693. https://doi.org/10.1130/G52976.1.

Sifting Volcanic Paydirt To Help Forecast Eruptions (Ned Rozell),  Alaska Science Forum, 3 October 2019.  Reposted in the AGU Blogosphere "The Field" on 7 October 2019.