Supervolcanoes: what monitoring can predict

Earthquakes, ground deformation and gas emissions help volcanologists monitor large volcanic systems. But detecting unusual activity is not the same as dating a super-eruption.

A volcano can tremble, swell or alter its gas emissions without a major eruption being imminent. For scientists, forecasting an eruption is therefore not a matter of reading a single sign as a countdown: they have to monitor several measurements, compare them with the volcano’s usual activity and determine whether they point to a coherent pattern of change.

This distinction is particularly important for large volcanic systems such as Yellowstone. Monitoring can identify changes and help assess a hazard. On its own, it cannot provide the date and size of a future eruption.

“Supervolcano” describes a past eruption

The term “supervolcano” refers to a volcanic centre that has produced at least one eruption of magnitude 8 on the Volcanic Explosivity Index, or VEI. For this category, the essential criterion is a volume of deposits exceeding 1,000 cubic kilometres. It is a classification based on the scale of a past eruption, not a diagnosis of what the volcano is about to do.

Yellowstone, Toba in Indonesia, Taupo in New Zealand and Long Valley in California are among the systems associated with ancient large explosive eruptions. Their calderas—vast depressions formed when the ground collapses after magma withdraws—bear the traces of these episodes. But a volcano that has experienced a super-eruption can also produce much more modest events.

Yellowstone offers a concrete example. According to the US Geological Survey (USGS), its most recent volcanic activity was a lava flow, about 70,000 years ago. The agency notes that, since the last major caldera eruption, about 640,000 years ago, the system has experienced numerous relatively non-explosive eruptions, mainly lava flows. The category “supervolcano” therefore means neither that every renewed spell of activity would be gigantic nor that such an eruption is due at regular intervals.

Different instruments monitor the same system

Volcanologists look for measurable changes in several aspects of a volcano. Seismology detects earthquakes, including swarms—that is, groups of closely spaced tremors. Their development can help identify rock fracturing or the movement of fluids at depth. But earthquakes are frequent in some volcanic regions: their presence alone is not enough to signal an eruption.

Geodesy measures ground deformation. GPS stations and radar satellites can track uplift or subsidence, sometimes down to a few millimetres. These variations provide information about changes in pressure and movements below the surface. Here too, ground deformation is not, on its own, proof that an eruption is in the making: volcanic systems also undergo fluctuations that do not lead to an eruption.

Scientists also monitor gas emissions, such as carbon dioxide and sulphur dioxide, as well as temperature and certain properties of water and hot springs. These measurements document the exchanges between magma, rocks, gases and groundwater. At Yellowstone, the Yellowstone Volcano Observatory, in which the USGS participates, monitors earthquakes, ground deformation, rivers and certain temperatures, among other things.

The value lies in bringing the evidence together. Unusual changes occurring simultaneously across several signals, observed in several locations and over time, are more informative than a single measurement taken alone. Specialists compare them with the baseline—in other words, the system’s usual activity. The USGS stresses that low-magnitude earthquakes, uplift or subsidence, and gas emissions are common phenomena at Yellowstone and do not, by themselves, signal an imminent eruption.

Detecting a crisis is not the same as dating an eruption

Three questions need to be distinguished. The first is whether a volcano’s activity is changing. The second is whether that change corresponds to a crisis that could lead to an eruption. The third, and more difficult, question is forecasting the event’s location, timing and scale. Instruments provide observations that contribute to this assessment; they do not turn a volcano into a clock.

An increase in activity may herald an eruption, but it may also subside or develop without producing the expected scenario. And the type of event matters: a lava flow and a major explosive eruption have different dynamics and consequences. The signals being sought and their interpretation must therefore be considered in relation to the volcano in question and the event being envisaged.

The USGS considers that an episode preceding a catastrophic eruption at Yellowstone would probably be detectable, but its exact duration cannot be determined in advance. The agency refers to preparatory activity that could appear weeks, or even months or years, beforehand. For smaller eruptions, some precursors may appear in the days or weeks before the event. These are general indications, not a guarantee of a timescale applicable to every volcano or every eruption.

Yellowstone is not “overdue”

Comparing the dates of past eruptions can create the illusion of a timetable. Yellowstone has experienced three major caldera eruptions, separated by very different lengths of time. Calculating an average between these intervals does not make it possible to infer the next date: two intervals are not enough to establish periodicity, and volcanic eruptions do not follow a regular rhythm.

The USGS also notes that Yellowstone’s shallow rhyolitic magma chamber is mostly solid, with only a fraction of its material molten. Scientists are not certain that there is enough mobile magma to fuel another major eruption. A catastrophic eruption remains possible in geological terms, but the agency says there is no sign that one is imminent and that such an event is unlikely in the next few centuries.

Monitoring does not promise to predict the day of a super-eruption. Its purpose is to observe changes, distinguish an ordinary fluctuation from a more concerning crisis, and communicate assessments as data come in. For readers, the key point is simple: an active volcano is not necessarily about to erupt, and an earthquake or an isolated measurement is not, on its own, enough to establish that an eruption is imminent.