The recent strong densification of seismic networks, and the increase of data availability from different type of sensors/networks (low-cost, Broadband and Strong-motion sensors, temporary deployments with hundreds of nodes, etc), questions our ability to automatically and accurately detect, locate and characterize seismic events combining these composite and large datasets. These steps are however key issues to monitor natural and anthropogenic seismic activity. As an example of what can be done with respect to this issue, I will first present:
1) the results of a recent study on an unusual seismic swarm occurring in the Pyrenees, where the use of both deep-learning and template-matching methods allowed to highlight new tectonic structures and a slow migration of the seismicity related to natural fluid-induced processes; and 2) a short focus on two others studied sites involving natural and induced seismicity (Arette region and Lacq gas-reservoir region), better apprehended thanks to the use of multiple deep-learning technics.
