The data-driven models developed by the team could become part of early warning systems used to predict future eruptions, with the potential to save lives and prevent damage to critical infrastructure.
University of Canterbury (UC) Civil and Natural Resources Engineering Research Engineer Dr Alberto Ardid and Associate Professor David Dempsey have used machine learning to analyse seismic patterns leading up to 41 previous eruptions across 24 volcanoes, including three in New Zealand.
They found these eruption warning signals follow repeatable patterns that can be transferred to other, less well-studied volcanoes.
"This finding could be a breakthrough for eruption forecasting, allowing us to use data from well-monitored volcanoes to improve monitoring and risk mitigation at under-monitored sites, enhancing volcano safety globally," Dr Ardid says.
Eruptions pose a significant threat to the 29 million people around the world who live within 10km of active volcanoes.
"Timely and accurate eruption forecasting can save lives, reduce economic losses, and minimise losses due to disruptions to air travel, agriculture, and global supply chains," Dr Ardid says. "Our method provides a cost-effective and scalable solution for improving forecasting at under-monitored volcanoes, benefiting communities and disaster management agencies globally."
Dr Ardid recently won the New Zealand Geophysics Prize for this research, which was supported by grants from the Ministry of Business, Innovation and Employment's Endeavour Fund, and from international partners. He says it's exciting to develop a tool that can contribute to volcano warning systems that prevent loss of life.
"It will be particularly valuable in developing countries where data is scarce, such as Southeast Asia and Central America, and that is a big motivation behind this project."
Associate Professor Dempsey says the research team has collaborated closely with volcano observatories internationally to ensure the new models provide actionable data. The plan is for these codes to be shared with volcano observatories here and overseas in an open-access policy.
"The modelling tool we've come up with is relatively simple and it's complementary to existing practices of volcanic observations, but it provides an extra layer of information," he says. "It means we can start to think about forecasting eruptions at volcanoes that have never had instrumentally recorded eruptions, such as Mount Taranaki," he says.
UC School of Earth and Environment volcanologist Professor Ben Kennedy, who also collaborated on the study published in Nature Communications today, says active volcanoes, such as Whakaari, Ruapehu, and Tongariro in Aotearoa New Zealand, are unpredictable and sometimes hazardous, but effective warning systems can help save lives and avoid debilitating injuries.
"This new research is really exciting because it challenges the current paradigm that eruption precursors, or warning signs, are unique to individual volcanoes.
"This is the first time we've had a model that demonstrates how we can use eruption precursor data from a number of volcanoes to help forecast a future eruption at another volcano where there's very little available data."
The research project was a collaboration with University of Auckland Professor Shane Cronin, along with 18 international researchers from nine countries.
