Abstract Ionospheric disturbances caused by acoustic waves (AW) are observed by Global Navigation Satellite Systems (GNSS) measures of total electron content (TEC). AWs that propagate to the ionosphere can be caused by ground‐based explosions, including seismic events. These events often occur in remote regions like the open ocean. Identifying an AW and locating the source of the AW from GNSS TEC measurements is challenging due to several variables between events such as propagation metrics, line of sight geometry, and ionization levels. This becomes especially difficult with a sparse network and limited signal coverage, typical of remote regions. We propose here the use of a deep learning, ionospheric anomaly detection algorithm and a geolocation algorithm capable of identifying atypical ionospheric behavior after a seismic event and estimating the source location. We outline the development of this algorithm and test results over the months of May and June 2023 for five GNSS receivers in the South Pacific. Results show the successful detection of two open ocean M7.7 and M7.1 earthquakes. Three TEC measurements detected as anomalous from the M7.7 earthquake event were used to estimate the epicenter of the earthquake, resulting in a source geolocation of S, E, a discrepancy of in latitude and in longitude, . This research demonstrates the feasibility of AW detection and source geolocation from a sparse GNSS network over open ocean.
Luhrmann et al. (Thu,) studied this question.