Session

DescriptionComputational geosciences leverage advanced computational methods to improve our understanding of the interiors of Earth and other planets. They combine numerical models to understand the current state of physical quantities describing a system, to predict their future states, and to infer unknown parameters of those models from data measurements. Such models produce highly nonlinear numerical systems with extremely large numbers of unknowns.
The ever-increasing power and availability of High Performance Computing (HPC) facilities offers researchers unprecedented opportunities to continually increase both the spatiotemporal resolution and the physical complexity of their numerical models. However, this requires complex numerical methods and their implementations that can harness the HPC resources efficiently for problem sizes of billions of degrees of freedom.
The goal of this minisymposium is to bring scientists who work in theory, numerical methods, algorithms and scientific software engineering for scalable numerical modelling and inversion. Examples include, but are not limited to, geodynamics, multi-phase geophysical flow modelling, seismic wave propagation and imaging, seismic tomography and inversion of large data sets, development of elaborate workflows including HPC for imaging problems, ice-sheet modelling, and urgent computing for natural hazards.