The Roof Sets the Size

01The evidence

• Tectonic stress has been shown to play a role in triggering large silicic caldera eruptions, with the regional stress field modulating whether a chamber fails at all and when.Geophysical Research Letters, 2018
• Caldera collapse has been modelled as a mechanical failure of the chamber roof under pressure, with the dynamics of collapse controlling the eruption's intensity rather than the other way round.JGR Solid Earth, 2026

02The tension

Recharge models explain unrest well and eruption timing poorly. Many restless calderas with large melt bodies never erupt, and some erupt with modest recharge. The field resolves this by invoking a threshold, but the threshold is fitted rather than predicted, which leaves the size of the eruption unforecast.

03The theory

A silicic chamber accumulates magma over hundreds of thousands of years. The magma is buoyant, so it pushes on the roof continuously. The roof is a plate of rock with a strength that depends on its thickness, its fracture state and the regional stress field.

Failure happens when the buoyancy pressure exceeds the roof's strength. The volume released at that moment is the volume that has accumulated, which is set by how long the roof held, which is itself set by the roof's properties. Recharge rate enters only as a slow clock.

The prediction is therefore that eruption size should correlate with roof geometry and regional stress, and only weakly with measured recharge rate. Calderas with thin or fractured roofs should erupt smaller volumes more often, and those with thick intact roofs should erupt rarely and enormously.

This also explains why forecasting is hard: the observable is magma, the controlling variable is rock. A monitoring programme that tracks inflation will see the clock but not the strength, and the two calderas with identical inflation histories can have very different futures.

04What this is not

05Predictions

• P1Eruption volume should correlate with roof thickness and fracture state better than with measured recharge rate.
• P2Calderas with thin or faulted roofs should show more frequent, smaller eruptions than thick-roofed systems at equal magma supply.
• P3Regional stress changes, such as those from nearby earthquakes, should shift eruption timing without changing the volume.

06What would refute it

Eruption size tracking magma supply across well-characterised calderas, with roof properties adding nothing, would return the problem to recharge.

07What it would change

It would move caldera monitoring from magma budgeting to roof imaging, and make eruption size a forecast rather than a surprise.