Forecasting volcanic eruptions and assessing their hazards is a central motivation for the field of volcanology. Whether for a frequently erupting volcano like Kīlauea in Hawaii or a restless caldera like Campi Flegrei in Italy, forecasting relies on making measurements in near-real-time to track changes in magmatic and volcanic activity. However, interpreting these signals is difficult without understanding how they relate to magmatic plumbing systems that cannot be directly observed. A trio of research papers in this issue demonstrate that petrology is key to understanding what geochemical and geophysical signals of restless volcanoes are telling us about the hazards they pose.

Credit: Dario Lopez-Mills / Associated Press / Alamy Stock Photo

Volcanologists use a range of complementary approaches to monitor volcanoes. For example, ground deformation measurements using satellite navigation systems and radar data can sense deformation induced by magmatic fluids; seismometers record diverse seismic signals generated by volcanic processes, such as rock fracturing and gas flow, and seismic data can be used to infer the location of magma at depth; and airborne and ground-based instruments track both the flux and composition of emitted gases.

Monitoring data is most reliably interpreted following long-term surveillance across eruptive cycles so that anomalous activity can be identified from background unrest. Hence maintaining funding of volcano observatories that conduct much of this monitoring is especially critical. For example, as Bruno Scaillet and colleagues explain in a News & Views, research on Campi Flegrei presented in this issue was made possible because of the continuous operation of the Vesuvius Observatory since 1841, which provided valuable context in interpreting monitoring data from the most recent period of unrest.

However, none of this monitoring, even if sustained at high spatial and temporal resolution and employing state-of-the-art instrumentation, allows a direct view of the underlying processes. Through petrology — analysing the mineralogical and chemical composition of sampled solidified volcanic products — the pulse of a volcano as measured by geophysical and geochemical measurements can be related to the anatomy of the magmatic system beneath1.

This brings us back to Campi Flegrei, the largest active caldera volcano in Europe where the most recent period of unrest since 2005 has been characterized by ground deformation and seismic activity2. Whether this unrest is caused by magmatic or hydrothermal activity has been debated. In an Article in this issue, Caliro et al. analyse a record of gas emission measurements spanning four decades. They identify a change in gas sulfur composition since 2018, but only by integrating their results with petrological evidence from melt inclusions are they able to attribute this sulfur signal to magmatic degassing.

Likewise, an Article by Pang et al. relies on prior petrology work to support their interpretation of the upper-crustal low-seismic-velocity zones they find beneath Cascade Range volcanoes in the northwest USA as magma bodies that persist at shallow depth throughout eruption cycles.

Petrology can provide critical constraints, but is usually limited to few data points compared to some other forms of monitoring. Petrological monitoring, which involves analysing volcanic products in near-real-time, has recently become possible. Not only can petrological monitoring track how magma properties change through an eruption, but its integration with data from other monitoring techniques can link these signals to pre-eruptive magma dynamics1. An Article by Longpré et al. make a case for the value of high-temporal-resolution petrological monitoring for assessing volcanic hazards in near-real-time. Their approximately daily sampling and compositional analysis of ash during the 2021 Tajogaite eruption of Cumbre Vieja volcano on La Palma, Canary Islands, reveals that the chemical composition of the erupted melt changes through the eruption and that this is linked to volcanic tremor generation.

There are other examples of petrological monitoring’s value: high-temporal-resolution analysis of volcanic ash from the 2021 Tajogaite eruption has been used to identify distinct phases of melt feeding the eruption3. Elsewhere, high-resolution time series of the isotopic composition of ash samples through the 1999–2016 eruption sequence of Tungurahua volcano in Ecuador has suggested that magma production there is regulated by chemical alteration by fluids4.

Eruption forecasting benefits from long-term and high-temporal-resolution monitoring of volcanoes. Petrology provides critical context to decode the gaseous whiffs and seismic trembles recorded by monitoring to improve understanding of volcanic processes and hazards. There are logistical challenges to near-real-time sampling of volcanic materials, requiring fieldwork at short notice for an unclear duration. Petrological monitoring also requires rapid execution of relatively time-consuming methods1. Nonetheless, petrology — and the direct window it offers to the magma beneath — should be an integral part of any volcano monitoring effort.