Séminaire ISTerre

---

Improved Earthquake Monitoring with AI

mardi 8 février 2022 - 11h00

Greg Beroza - Stanford University
--- 
There is a well-established workflow for earthquake monitoring that includes a sequence of steps that include phase detection, phase association, location, and characterization that is used to develop seismicity catalogs around the world and across scales. Because the number of earthquakes is universally observed to increase rapidly as magnitude decreases, cataloging smaller earthquakes will dramatically increase the information available to study earthquake processes. In the past decade, the methods of AI – initially data mining, and more recently machine learning – have been applied to seismic monitoring to great effect.  Appropriate architectures, accurate data labels, and data augmentation all play important roles in developing effective models that generalize well. The simplest approach is modular in which individual earthquake monitoring tasks are replaced one-by-one with neural network models that are applied in serial.  There should be advantages, however, in combining steps in multi-task models, with an end-to-end model as an extreme end member, to take advantage of contextual information.  AI-based earthquake monitoring is now being deployed for real-time monitoring, and there is no reason for it not to be applied comprehensively to available archived data.  Seismologists are now developing catalogs for both natural and induced seismicity that are far more comprehensive than before that often feature an order of magnitude more small earthquakes.  The next challenge will be to use this clearer view of  seismicity to understand better the mechanics of earthquake processes.  The methods of AI should also be useful for this effort.

---

Equipe organisatrice : Ondes et structures

Séminaire uniquement en visio

Informations de visio :

https://univ-grenoble-alpes-fr.zoom.us/j/96913102209?pwd=aCthMFNEV0lmRGxqd3h5K3p5QWpHZz09