How Eruption Forecasting Works

March 23, 2026

Earthquakes cannot be predicted. Eruptions often can be — not perfectly, but well enough to evacuate cities. The reason is simple: a fault can slip in an instant, but a volcano cannot erupt without first moving a large mass of magma toward the surface, and that movement is loud.

The signs of rising magma

Pinatubo, 1991: the model case

Pinatubo in the Philippines had not erupted in 500 years and was not monitored. In early April 1991 it began producing steam explosions and earthquakes. A joint team from the Philippine volcanology institute and the USGS installed instruments, tracked escalating seismicity and gas, and raised the alert level in stages.

They evacuated in widening rings as the signs worsened — more than 60,000 people by 14 June. On 15 June the volcano produced a VEI 6 eruption that buried the surrounding area in ash and pyroclastic flows. The forecast is credited with saving thousands of lives and hundreds of millions of dollars in aircraft alone.

Where it still fails

Forecasting is not prediction. Observatories can usually say that an eruption is likely within days to weeks, but the size is hard to call in advance, and unrest sometimes fades without any eruption at all — a “failed” forecast that still costs a real evacuation.

The bigger gap is coverage. Of the roughly 1,300 potentially active volcanoes on the map, only a few hundred have real ground-based monitoring. Volcanoes that erupt rarely are the least studied, and some of the worst disasters — Nevado del Ruiz in 1985, which killed 23,000 despite an accurate hazard map — came down to warnings that existed but did not reach or convince the people in the path.

On the map, the “everything up to here” view shows which volcanoes have erupted in the last century. A volcano with no eruption since 1900 is not necessarily safe — it may simply have a long repose interval, like Pinatubo did.

Sources

  1. U.S. Geological Survey — Volcano Hazards Program: “Monitoring volcanoes”, “Forecasting eruptions”, “Pinatubo 1991”.
  2. Newhall, C. G. & Punongbayan, R. S. (eds.) (1996). Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines.
  3. Sparks, R. S. J. (2003). “Forecasting volcanic eruptions.” Earth and Planetary Science Letters 210, 1–15.
  4. Voight, B. (1990). “The 1985 Nevado del Ruiz volcano catastrophe.” J. Volcanol. Geotherm. Res. 44, 349–386.
← Explore the Interactive Map