Deep learning forecasting of large induced earthquakes via precursory signals.

Precursory phenomena to earthquakes have always attracted researchers' attention. Among the most investigated precursors, foreshocks play a key role. However, their prompt identification with respect to background seismicity still remains an issue. The task is worsened when dealing with low-magnitude earthquakes. Despite that, seismology and, in particular real-time seismology, can nowadays benefit from the use of Artificial Intelligence (AI) to face the challenge of effective precursory signals discrimination. Here, we propose a deep learning method named PreD-Net (precursor detection network) to address precursory signal identification of induced earthquakes. PreD-Net has been trained on data related to three different induced seismicity areas, namely The Geysers, located in California, USA, Cooper Basin, Australia, Hengill in Iceland. The network shows a suitable model generalization, providing considerable results on samples that were not used during the network training phase of all the sites. Tests on related samples of induced large events, with the addition of data collected from the Basel catalogue, Switzerland, assess the possibility of building a real-time warning strategy to be used to avoid adverse consequences during field operations.

Imaging overpressurised fracture networks and geological barriers hindering fluid migrations across a slow-deformation seismic gap.

There is an ongoing debate on the processes producing background seismicity and deformation transients across seismic gaps, i.e., regions that lack historical large-magnitude earthquakes. Essential missing elements are geophysical images that resolve sources of geophysical unrest. Here, we apply seismic scattering and absorption tomography to data recorded during the 2010-2014 seismic sequence within the Mt. Pollino seismic gap region (Southern Italy). The tomographic models show high sensitivity to fluid content, deformed fractured structures, and impermeable layers stopping fluid migrations. They bridge the gaps between geological and geophysical models and provide a highly-resolved image of the source of seismic and deformation unrest within this seismic gap. High absorption topping the western Pollino seismic volume appears pressurized between the low-Vp/Vs and low-scattering San Donato metamorphic core and a deep basement. Absorbing fluids can only migrate laterally to the east, blocked in the west and southwest by deep low-scattering barriers associated with east-dipping faults and to the north and southeast by saturated overpressurized low-scattering basins. This eastern migration is only partially effective, producing seismicity across the lowest boundary of the high-absorption volume. Our results showcase the potential of seismic scattering and absorption when imaging structures causing geophysical unrest processes across fault networks.

Ground motion prediction equations as a proxy for medium properties variation due to geothermal resources exploitation.

Sub surface operations for energy production such as gas storage, fluid injection or hydraulic fracking modify the physical properties of the crust, in particular seismic velocity and anelastic attenuation. Continuously measuring these properties may be crucial to monitor the status of the reservoir. Here we propose a not usual use of the empirical ground-motion prediction equations (GMPEs) to monitor large-scale medium properties variations in a reservoir during fluid injection experiments. In practice, peak-ground velocities recorded during field operations are used to update the coefficients of a reference GMPE whose variation can be physically interpreted in terms of anelastic attenuation and seismic velocity. We apply the technique to earthquakes recorded at The Geysers geothermal field in Southern California and events occurred in the St. Gallen (Switzerland) geothermal field. Our results suggest that the GMPEs can be effectively used as a proxy for some reservoir properties variation by using induced earthquakes recorded at relatively dense networks.

3D ultra-high resolution seismic imaging of shallow Solfatara crater in Campi Flegrei (Italy): New insights on deep hydrothermal fluid circulation processes.

Seismic tomography can be used to image the spatial variation of rock properties within complex geological media such as volcanoes. Solfatara is a volcano located within the Campi Flegrei, a still active caldera, so it is of major importance to characterize its level of activity and potential danger. In this light, a 3D tomographic high-resolution P-wave velocity image of the shallow central part of Solfatara crater is obtained using first arrival times and a multiscale approach. The retrieved images, integrated with the resistivity section and temperature and the CO2 flux measurements, define the following characteristics: 1. A depth-dependent P-wave velocity layer down to 14 m, with Vp < 700 m/s typical of poorly-consolidated tephra and affected by CO2 degassing; 2. An intermediate layer, deepening towards the mineralized liquid-saturated area (Fangaia), interpreted as permeable deposits saturated with condensed water; 3. A deep, confined high velocity anomaly associated with a CO 2 reservoir. These features are expression of an area located between the Fangaia, water saturated and replenished from deep aquifers, and the main fumaroles, superficial relief of the deep rising CO2 flux. Therefore, the changes in the outgassing rate greatly affect the shallow hydrothermal system, which can be used as a "mirror" of fluid migration processes occurring at depth.