In late 2017, researchers implemented a new volcano forecasting modelling tool on both the Blue Waters and iForge supercomputers. Another crew was keeping a watch on Ecuador’s Galapagos Islands’ Sierra Negra volcano. Because of a connection discovered by one of the scientists working on the Ecuador study, they were able to make a lucky five-month prognosis of the Sierra Negra eruption in June 2018.
The groundbreaking modelling approach, which originated on an Apple iMac computer, has already gotten a lot of press for its ability to accurately replicate the 2008 eruption of Alaska’s Okmok volcano. Scientists at the National Center for Supercomputing Applications wanted to put the model’s new high-performance computing upgrade to the test, and they indicated that an eruption was imminent.
The Sierra Negra volcano, according to Patricia Gregg, the principal author of a new paper proving the initiative’s effectiveness, is a well-behaved volcano. Prior to past eruptions, the volcano showed all of the warning signs of an impending eruption, including ground tremors, gas leaks, and increased seismic activity. Sierra Negra is an outstanding choice to serve as a test subject for our revised model because of this attribute.
However, a significant number of volcanoes do not follow these well-established patterns. Predicting volcanic eruptions is one of the most difficult aspects of volcanology. The research of Gregg and her colleagues focuses on the construction of quantitative models that could aid in the interpretation of more complex events.
The perfect prediction.
During the winter break between 2017 and 2018, Gregg and her team used the newly created high-powered computer model to process the data from Sierra Negra. Despite the fact that the run was intended to be a test, it ended up providing a foundation for comprehending Sierra Negra’s eruption cycles, as well as assessing the likelihood of future eruptions and their timing, despite the fact that no one realised it at the time.
The model predicted that the strength of the rocks that house Sierra Negra’s magma chamber were deteriorating, This was predicted to result in a mechanical failure and ultimately eruption. In March of 2018, the team presented the findings from our inquiry at a conference of scientists. Due to mechanical failure, the Sierra Negra volcano erupted one day later than expected.
Despite the fact that the study’s setting is ideal, the researchers believe the findings demonstrate the need for high-performance supercomputing in real-world research. The advantage of this upgraded model is its capacity to continually collect multimodal, real-time data and quickly process it to offer a day forecast akin to weather forecasting. The model’s ability to ingest diverse real-time input and analyse it rapidly enough to create a daily forecast has been improved. This necessitates a vast amount of processing power that the volcanology community lacked earlier.
Gregg’s team did not have access to a methodology that merged various disciplines prior to working with NCSA. They needed to integrate the components in this way to put together a modelling application of this calibre.
The future of predicting volcanic eruptions.
The team plans to include machine learning and artificial intelligence into the forecasting model, allowing researchers to access this level of processing capacity using only basic laptops and desktop computers.
The study was recently published as an article in Science Advances.
Story Source: Original press release by University of Illinois at Urbana-Champaign, News Bureau.. Note: Content may be edited for style and length by Scible News.
References
Patricia M. Gregg, Yan Zhan, Falk Amelung, Dennis Geist, Patricia Mothes, Seid Koric, Zhang Yunjun. Forecasting mechanical failure and the 26 June 2018 eruption of Sierra Negra Volcano, Galápagos, Ecuador. Science Advances, 2022; 8 (22) DOI: 10.1126/sciadv.abm4261
