"Earthquakes represent one of the deadliest and costliest natural disasters affecting our planet – and one of the hardest to predict. To improve seismic hazard evaluation in earthquake-prone regions, an understanding of earthquake nucleation and of the underlying microphysical and chemical processes is crucial. A better understanding of the processes that control earthquake nucleation is also of rapidly growing importance for mitigation of induced seismicity, caused by activities such as gas and oil production, and geological storage of CO2 or gas. The SEISMIC project is a multi-scale study aimed at understanding the parameters that control slip (in)stability in experiments and models addressing earthquake nucleation. A central question to be tackled is what controls the velocity-dependence of fault friction and hence the potential for accelerating, seismogenic slip, and on what length scales the processes operate. A novel acoustic imaging technique will be developed and applied in experiments to obtain direct information on the internal microstructural evolution of fault slip zones during deformation, and on how this evolution leads to unstable slip. The SEISMIC project will link experiments with sophisticated numerical models of grain-scale frictional processes. Using both experiments and grain scale modelling, the SEISMIC project will in turn directly test boundary element models for large scale fault slip. The coupling of experiments with grain-scale numerical models, based on in-situ imaging, will provide the first, integrated, multiscale physical basis for extrapolation and upscaling of lab friction parameters to natural conditions. Ultimately, the SEISMIC project will test and validate the resulting models for fault slip by simulating and comparing patterns of seismicity for two natural-laboratory cases: a) for the l’Aquila region of Central Italy, and b) for a reservoir-scale case study involving induced seismicity in the Netherlands."