A physics-based numerical approach is used to characterize earthquake ground motion due to induced seismicity in the Groningen gas field and to improve empirical ground motion models for seismic hazard and risk assessment. To this end, a large-scale (20 km × 20 km) heterogeneous 3D seismic wave propagation model for the Groningen area is constructed, based on the significant bulk of available geological, geophysical, geotechnical, and seismological data. Results of physics-based numerical simulations are validated against the ground motion recordings of the January 8, 2018, M L 3.4 Zeerijp earthquake. Taking advantage of suitable models of slip time functions at the seismic source and of the detailed geophysical model, the numerical simulations are found to reproduce accurately the observed features of ground motions at epicentral distances less than 10 km, in a broad frequency range, up to about 8 Hz. A sensitivity analysis is also addressed to discuss the impact of 3D underground geological features, the stochastic variability of seismic velocities and the frequency dependence of the quality factor. Amongst others, results point out some key features related to 3D seismic wave propagation, such as the magnitude and distance dependence of site amplification functions, that may be relevant to the improvement of the empirical models for earthquake ground motion prediction.

Modelling natural composites, as the majority of real geomaterials, requires facing their intrinsic multiscale nature. This allows to consider multiphysics coupling occurring at the microscale, then reflected onto the macroscopic behavior. Geotechnics is constantly requiring reliable constitutive models of natural composites to solve large-scale engineering problems accurately and efficiently. This need motivates the contribution. To capture in detail the macroscopic effects of microscopic processes, many authors have developed multi-scale numerical schemes. A common drawback of such methods is the prohibitive computational cost. Recently, Machine Learning based approaches have raised as promising alternatives to traditional methods. Artificial Neural Networks -ANNs -have been used to predict the constitutive behaviour of complex, heterogeneous materials, with reduced calculation costs. However, a major weakness of ANN is the lack of a rigorous framework based on principles of physics. This often implies a limited capability to extrapolate values ranging outside the training set and the need of large, high-quality datasets, on which performing the training. This work focuses on the use of Thermodynamics-based Artificial Neural Networks -TANN -to predict the constitutive behaviour of natural composites. Dimensionality reduction techniques -DRTs -are used to embed information of microscopic processes into a lower dimensional manifold. The obtained set of variables is used to characterize the state of the material at the macroscopic scale. Entanglement of DRTs with TANN allows to reproduce the complex nonlinear material response with reduced computational costs and guarantying thermodynamic admissibility. To demonstrate the method capabilities an application to a heterogeneous material model is presented.