Subduction Zones in Four Dimensions (SZ4D) is launching TEFRA (Testing Eruption Forecasting through Research and Assimilation), a new project to establish the first open, community-scale competition in volcanic eruption forecasting. TEFRA will be supported by the Gordon and Betty Moore Foundation.

Learn more | Visit the TEFRA website

Apply to join us for a full-day, in-person meeting preceding the 2026 AGU Fall Meeting to help shape the TEFRA competition and begin building the research teams that will participate in it.

The TEFRA (Testing Eruption Forecasting through Research and Assimilation) project will establish the first open, community-scale competition in volcanic eruption forecasting, designed to rigorously produce and evaluate competing approaches to eruption prediction using modern, open-access datasets.

TEFRA is supported by the Gordon and Betty Moore Foundation.

The project includes both the infrastructure for the competition and funds for research teams to develop the models. Funds will be available for teams to do the necessary research based on a proposal call in Northern hemisphere Spring 2027. The competition and research team funding will run from approximately September 2027 through June 2029. The purpose of the TEFRA Launch Meeting at AGU26 is to initiate conversations about volcano targets, forecast models and outcomes that will guide the competition framing and seed the formation of research teams. 

Although volcanic eruptions are among the few natural hazards that have proven potential for successful forecasts, forecasting today remains highly dependent on expert interpretation and volcano-specific experience rather than reproducible, quantitative scientific frameworks. All but a few of the hundreds of eruptions over the last decades have occurred without a quantitative forecast (e.g., a 50% chance of eruption in the next two weeks). We aim to transform this landscape by using the Subduction Zones in Four Dimensions (SZ4D) network of scientists to organize an international forecasting competition in which independent teams will generate and apply statistical, physics-based, and hybrid AI/data-assimilation methods to a common suite of target volcanoes and standardized datasets. By directly comparing forecasting strategies against the same eruptions, observational constraints, and evaluation metrics, TEFRA creates comparable models that can determine where approaches succeed or fail, and why.

The competition structure is the scientific core of the project. Teams will develop and test competing forecasting models on a carefully selected suite of well-observed volcanic systems spanning contrasting eruption styles and forecasting challenges, including closed-vent, open-vent and hydrothermal systems, which are known to have different eruptive styles and precursors. Forecasts will be evaluated quantitatively and compared to a benchmark model on the accuracy of the eruption onset, eruption magnitude, intensity, style, vent location and duration. Annual workshops open to the broad community and coordinated benchmarking exercises will ensure iterative comparison, refinement, and cross-fertilization between teams while maintaining independent model development. The result will be a suite of operational volcano forecasting models that have been tested retrospectively and are thus ready to be put to the test of true prospective prediction.

The launch meeting will bring together volcanology, forecasting and computation experts with potential competition participants and end-users to develop guidelines for the competition that will maximize the utility of the final forecasting models. (See application for more description of relevant expertise.) The discussions of this launch meeting will inform the research team funding call expected in the 2nd quarter of 2027.

The major agenda items will be:

• Discussion of target volcanoes

The criteria for volcano selection are:

• Open and accessible high-quality, multi-parameter observations, including at minimum seismic and geodetic data, and preferably gas, thermal, and infrasound data available. All datasets must be entirely open and available on a public website to ensure reproducible analyses and an open competition. 

• Well-documented eruptive sequences with baseline data, unrest, escalation, and eruption timelines.

• Representation of contrasting conduit states and eruption styles, including open-vent, closed-system, and hydrothermal/phreatic eruption.

• Relevance to subduction zone hazards and broader MultiHazardArray science goals.

In the TEFRA proposal, the following volcanoes were provided as exemplars:

• Okmok 2008 | closed, caldera

• Semisopochnoi 2021 | closed, clear escalating unrest, failed eruptions prior

• Villarrica 2015 | open vent, active lava lake

• Poás 2024-2025 | phreatomagmatic, crater lake

2. Discussion of forecast success criteria, timing  (e.g., 24h, 1 week, 1 month), and tiering.

Forecasts could predict eruption likelihood, onset, eruption magnitude, intensity, style, vent location and duration. The forecasts should be updateable as new data are available during an eruption. Crafting a quantitative method of evaluating success must be guided by the experience and needs of volcano monitoring observatories and agencies.

3. Discussion of potential model strategies.

AI, physical models, and statistical approaches are all possible. Teams are encouraged to incorporate a variety of techniques. The final model must be reproducible with the given data by an objective third party. All models will be open and publicly available.

The workshop will provide a venue to begin to discuss the benefits and drawbacks of various strategies in order to motivate fully informed research teams.

Participation

Please complete this form to express your interest in participating in the meeting.

Participation is free and does not require AGU registration. This is an in-person-only event.

Participation will be limited to 70 attendees. Applications will be chosen with the goal of assembling a mix of relevant expertise, perspectives, and potential contributions to the meeting.

Because we expect strong interest and space is limited, we encourage you to apply as early as possible. The application deadline is November 2, 2026. Applications will be evaluated on a rolling basis, and successful applicants will be notified no later than November 5.

Travel Support

Limited travel support is available for participants who need assistance. The application includes an opportunity to request travel support if necessary.

Important Dates

• Sept 15: Applications open

• Sept 15–Nov 2: Applications reviewed on a rolling basis until all spots are filled

• Nov 2: Application deadline

• Nov 5: Evaluation period ends; all successful applicants notified by this date

Questions

Contact the SZ4D Office at contact@sz4d.org