Detection of co-seismic ionospheric disturbances in near-real-time by using machine learning
Abstract
Tsunamis generated by large earthquake-induced displacements of the ocean floor can lead to tragic consequences for coastal communities. Ionospheric measurements of Co-Seismic Disturbances (CIDs) offer a unique solution to characterize an earthquake's tsunami potential in Near-Real-Time (NRT) since CIDs can be detected within 7-10 min of a seismic event. However, the detection of CIDs rely on human experts which currently prevents the deployment of ionospheric methods in NRT. To address this critical lack of automatic procedure, we build a Machine Learning (ML) model using random forests to (1) classify ionospheric waveforms between CIDs and noise, (2) pick arrival times, and (3) associate arrivals across a satellite network in NRT. The ML model was trained over an extensive dataset of GNSS-derived total electron content (TEC) signatures for 12 large-magnitude earthquakes and it shows excellent classification and arrival-time picking performances. Using automatically extracted CID arrival times, we further demonstrate how to apply seismo-ionospheric methods to locate regions of large surface slips and to localize the seismic source from ionospheric TEC measurements in the NRT scenario.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFMSA11B..01B