Supercomputing of large earthquakes

Abstract

<p>Earth modeling is extraordinarily complex - and when the earth quakes, it’s no different. Earthquakes are highly non-linear and multiscale processes fracturing the Earth’s crust and emanating potentially destructive seismic waves. While computational seismology has been a pioneering field for high-performance computing, the multitude of scales and multi-physics character of earthquake source processes remain difficult to constrain.</p><p>Earthquake science is increasingly data-rich, which opens up new pathways to synergistically integrate seismological, geodetic, tectonic and experimental analysis in multi-physics forward modeling. Using a physics-based description of earthquakes, interdisciplinary earthquake observations,  modern numerical methods and hardware specific optimisation shed light on the dynamics, severity and cascading hazards of earthquake behaviour. An unparalleled degree of realism is enabled by exploiting high-performance computing.</p><p>In this lecture, I will demonstrate the potential of Solid Earth community software for performing data-integrated, parsimonial scenarios of recent powerful multi-fault earthquake cascades, 3D fully-coupled Earth and ocean models of tsunami generated during earthquakes as well as cycles of earthquake and aseismic slip using petascale supercomputers. The inclusion of probabilistic and Bayesian frameworks, geometric transformations and diffuse interface approaches are future directions I will discuss for exploiting the expected exascale computing infrastructure.</p>