Seismic signals may predict volcanic eruption styles

Enlarge

Volcanic eruptions are always spectacular events. But whether or not an eruption poses an acute threat to human societies largely depends on its ‘style’. In many cases, phases of volcanic activity proceed in a relatively undramatic manner – for example, when hot lava of low viscosity, which contains only small amounts of gas and crystalline inclusions, seeps from fissures. At the other end of the scale are explosive eruptions, in which highly viscous lava, including large boulders, is suddenly expelled under high pressure from the volcano’s main vent, followed by tons of ash that can reach heights of several kilometers and/or pyroclastic clouds of glowing embers that race down the volcano’s flanks.

Generally speaking, the more explosive the eruption, the greater the risk for those who live nearby, for infrastructure and – in light of the deleterious effects of these eruptions on the climate – for society as a whole. This explains why volcanologists have long sought ways to predict the style of volcanic eruptions. An international team, including several researchers at LMU, has now taken a significant step towards the realization of this goal. In a new study, they have shown that a specific seismic signal that is detectable hours or even days before an eruption is directly correlated with the viscosity of the rising lava in the active vent. In other words, as it rises from the chamber, magma of low viscosity gives rise to seismic waves that can be distinguished from those associated with the presence of highly viscous material in the vent. “We hope that this will allow us to discriminate between explosive and less dangerous eruptions,” says Donald Dingwell, Director of the Department of Earth and Environmental Sciences at LMU.

The authors of the report, which appears in the journal Nature, studied a series of 24 eruptions of the Kīlauea volcano on the island of Hawaii that occurred in 2018. The geologists focused on the chemical and physical properties of the ascending magma, while the seismologists directed their attention to the patterns of seismic waves associated with these eruptions. “The eruption itself was full of unexpected twists, like the plot of a play,” Dingwell recalls. “Every eruption looked different from the previous one, and the style of the eruptions varied from the explosive to the relatively harmless end of the spectrum.” This pattern itself reveals that the chemical composition of magma in one and the same volcano can vary continuously with respect to features such as the relative contents of crystals, silicon dioxide and gas bubbles – and these are precisely the factors that determine the viscosity of magmas. “We were fortunate insofar as we had insisted that every eruption event should be investigated in detail,” Dingwell remarks.

A massive rotation of the stress field

Crucially, these high-precision experiments – carried out by Arianna Soldati, a Humboldt Fellow at LMU – allowed scientists from the Carnegie Institution of Science, the University of Hawaii and the United States Geological Survey to establish a reliable correlation between the physicochemical properties of the magmas and the seismological data. “This link between the viscosity of magma and seismicity is unprecedented,” says Dingwell.

The unexpected association between these variables became apparent when the researchers analyzed one specific seismic component called the fault-plane solution, which is a kind of tensor that marks the ‘first-motion polarity’ of a seismic source. Up to a certain level of viscosity of the magma, this parameter was oriented in a particular plane. As the viscosity of the magma increased, its orientation suddenly pivoted by 90 degrees. “This sudden change points to a massive rotation of the stress field, which is directly reflected in the seismic signal,” Dingwell explains. In principle, this signal offers a means of externally evaluating the potential level of risk presented by an upcoming eruption. With the aid of a suitably distributed seismic network positioned around particularly hazardous volcanos, this discovery may make it possible to warn those in the danger zone in good time prior to a pending eruption.

Dingwell plans to experimentally assess and characterize the situation that gives rise to the signal in his laboratory in Munich, with a view to understanding how the dynamics of the magma can change on such a short timescale. For the chemical and physical processes that underlie the sudden switch in the orientation of the fault-plane solution remain enigmatic.

Nature, 2021.