In 2018, a powerful earthquake hit Sulawesi in Indonesia. Palu – the capital of Central Sulawesi – was struck by an unexpected ensuing tsunami. In total, over 4,300 people were killed. Now, an international, interdisciplinary team of researchers has leveraged powerful supercomputing to better understand the mystery of the coupled disasters.
Larger tsunamis are usually caused by vertical motion in earthquakes. So when the earthquake struck and offset the ground horizontally, residents were told to expect a small (>0.5m) tsunami, if any. Instead, the tsunami carried waves in excess of 2m, sometimes reaching the second floors of buildings.
The satellite data suggested straight, smooth faults, so researchers ran through a series of possible explanations for the unexpected water displacement: landslides, offshore faults not seen by satellites, rupture speed, etc.
To answer this question, they turned to the SuperMUC phase 2 supercomputer at the Leibniz Supercomputing Center near Munich, Germany. SuperMUC phase 2, which was inaugurated in 2015, is an IBM System X iDataPlex with 86,016 cores across 6,144 Intel Xeon E5-2697 v3 processors. Phase 2 is independently capable of 3.6 petaflops of peak performance.
The researchers modeled the earthquake over the course of two and a half hours on 560 of phase 2’s Haswell cores. Crucially, the researchers’ model differed from traditional earthquake models – which attempt to fit the earthquake to the data – and instead used a more physics-driven model that incorporated physical processes along the fault. Using this model, they evaluated several available datasets.
The researchers found that the tsunami could, in fact, be explained by the earthquake causing movements in the seafloor of Palu Bay, and that landslides were not required to explain its severity. Specifically, the model indicated a very fast rupture on a tilted fault, causing primarily lateral – but also downward – motion. This, they concluded, resulted in a range of 0.8m to 2.8m vertical seafloor change (averaging 1.5m), likely accounting for the severe and unexpected height of the Palu tsunami.
“Finding that earthquake displacements probably played a critical role generating the Palu tsunami is as surprising as the very fast movements during the earthquake itself,” said Thomas Ulrich, PhD student at Ludwig Maximilian University of Munich and lead author of the paper. “We hope that our study will launch a much closer look on the tectonic settings and earthquake physics potentially favouring localized tsunamis in similar fault systems worldwide.”
About the research
The research referenced in this article was published as “Coupled, Physics-Based Modeling Reveals Earthquake Displacements are Critical to the 2018 Palu, Sulawesi Tsunami” in the August 2019 issue of Pure and Applied Geophysics. It was written by T. Ulrich, S. Vater, E.H. Madden, J. Behrens, Y. van Dinther, I. van Zelst, E. J. Fielding, C. Liang and A.-A. Gabriel.
To read the original article discussing the research posted by the Gauss Centre for Supercomputing, click here.