A comprehensive empirical study of T-wave properties

The Sound Fixing and Ranging (SOFAR) channel facilitates the propagation of underwater acoustic waves over vast distances with minimal attenuation. Exploiting this property, various passive hydroacoustic networks have been deployed globally to monitor signals trapped within the SOFAR channel. One notable application involves T-waves (or "Tertiary waves"), which are acoustic waves generated from the conversion of  seismic waves at the ocean-bottom interface and propagating in the SOFAR. T-waves offer significant advantages in seismic monitoring. For instance, hydrophones positioned strategically across the ocean often provide better earthquake detection coverage compared to terrestrial seismic stations, especially for low-magnitude events occurring remotely from land-based instruments, such as along mid-ocean ridges. The ability to detect earthquakes with magnitudes as low as ~3.0 (lower magnitude of completeness) enables us to focus on microseismicity that might otherwise go unnoticed.

Over the past few decades, numerous studies have used T-waves for monitoring seismic events. The detection of the same event at several hydrophone stations enables the time and origin of earthquakes to be estimated by least-squares inversion. Although these analyses have proved very useful, one wonders whether T-waves can tell us more about earthquakes. For example, some studies have established empirical relations between T-wave source level and earthquake magnitude to estimate seismic moment rates in remote parts of the ocean. Such relations, however, typically rely on very limited data sets (e.g., a few hundred events). Modeling work also suggests that additional T-wave parameters, such as amplitude and duration of the wave, or its frequency content, might reflect parameters such as rupture length or seafloor roughness. However, these hypotheses require thorough empirical validation, which has been hampered by the lack of comprehensive data sets.

To address this issue, we leverage a new detection tool called TiSSNet to automate the picking process of T-waves generated by events from the ISC ocean-wide catalog of teleseismic events. This provides a dataset of approximately 10,000 events, enabling extensive empirical comparisons focused on magnitude and source level parameters.