In the fall of 2017, geology professor Patricia Gregg and her team had just implemented a new volcanic prediction modeling program on Blue Waters and iForge supercomputers. Simultaneously, another team was monitoring the activity of the Sierra Negra volcano in the Galapagos Islands, Ecuador. One of the Ecuadorian project scientists, Dennis Geist of Colgate University, contacted Gregg, and what happened next was the chance prediction of the June 2018 Sierra Negra eruption five months before it not happen.
Originally developed on an iMac computer, the new modeling approach had already gained attention for successfully recreating the unexpected 2008 eruption of Okmok Volcano in Alaska. Gregg’s team, based at the University of Illinois at Urbana-Champaign and the National Center for Supercomputing Applications, wanted to test the model’s new high-performance computing upgrade, and Geist’s observations of the Sierra Negra showed signs of an impending blowout.
“Sierra Negra is a well-behaved volcano,” said Gregg, lead author of a new report on the successful effort. “This means that prior to past eruptions, the volcano showed all the telltale signs of an eruption that we would expect to see such as ground swell, gassing and increased seismic activity. This feature made the Sierra Negra a great test case for our upgraded model.”
However, many volcanoes do not follow these clearly established patterns, the researchers said. Predicting eruptions is one of the great challenges of volcanology, and developing quantitative models to help with these trickier scenarios is central to the work of Gregg and his team.
During the 2017-2018 winter break, Gregg and his colleagues analyzed data from Sierra Negra through the new supercomputing-powered model. They completed the run in January 2018, and although it was a test, it ended up providing a framework for understanding Sierra Negra eruption cycles and assessing the potential and timing of future eruptions. – but no one has yet noticed.
“Our model predicted that the strength of the rocks that contain the Sierra Negra magma chamber would become very unstable between June 25 and July 5, and possibly lead to mechanical failure and subsequent eruption,” Gregg said, also NCSA faculty member. . “We presented this conclusion at a scientific conference in March 2018. After that, we got busy with other work and didn’t see our models again until Dennis texted me on June 26, asking me to confirm the date we had planned.Sierra Negra erupted a day after our first scheduled mechanical failure date.We were floored.
Although this is an ideal scenario, according to the researchers, the study shows the power of integrating high-performance supercomputing into practical research. “The advantage of this improved model is its ability to continuously assimilate real-time multidisciplinary data and quickly process it to provide a daily forecast, similar to weather forecasts,” said Yan Zhan, a former graduate student from Illinois. and co-author of the study. “This takes an incredible amount of computing power previously unavailable to the volcanic forecasting community.”
Setting up the moving parts to produce a modeling program for this force requires a highly multidisciplinary approach that Gregg’s team did not have access to prior to working with NCSA.
“We all speak the same language when it comes to the numerical multiphysics analysis and high performance computing needed to predict mechanical failure – in this case of a volcanic magma chamber,” said Seid Koric, deputy technical director at NCSA, a research center professor of mechanical sciences and engineering and co-author of the study.
Using Koric’s expertise, the team said they hope to incorporate artificial intelligence and machine learning into the forecasting model to help make that computing power available to researchers working from standard laptop and desktop computers.
The results of the study are published in the journal Scientific advances.
Geist is program director at the National Science Foundation and professor of geology at Colgate University. Falk Amelung of the University of Miami; Patricia Mothes of the Instituto Geofísico Escuela Politecnica Nacional, Ecuador; and Zhang Yunjun from the California Institute of Technology also contributed to this research.
The National Science Foundation, NASA and NCSA supported this study.