Analysis of seafloor seismograms of the 2003 Tokachi-Oki earthquake sequence for earthquake early warning

Earthquake Early Warning (EEW) algorithms estimate the magnitude of an underway rupture from the first few seconds of the P-wave to allow hazard assessment and mitigation before the S-wave arrival. Many large subduction- zone earthquakes initiate 50­-150 km offshore, potentially allowing seafloor instruments sufficient time to identify large ruptures before the S-waves reach land. We tested an EEW algorithm using accelerograms recorded offshore Hokkaido in the region of the 2003 Mw 8.1 Tokachi-Oki earthquake and its aftershocks. A wavelet transform of the first 4 s of the P-wave concentrates information about earthquake magnitude from both waveform amplitude and frequency content. We find that wavelets with support of a few seconds provide discriminants for EEW that are both accurate enough to be useful and superior to peak acceleration or peak velocity. Additionally, we observe a scaling of wavelet coefficient magnitude above Mw 6.0 indicating that, at least for the mainshock (Mw 8.1) and largest aftershock (Mw 7.1), the final size of a rupture could have been estimated from the initial portion of the seismogram. Our work strengthens the arguments for establishing seafloor based EEW observatories in subduction zones, simply by the unique sensitivity of such data to earthquake source properties. Near-source seismograms are not contaminated with the arrivals of later phases, particularly refracted phases that can distort the waveform's frequency content. Thus, EEW metrics will be more accurately determined from local ocean bottom seismometer (OBS) data than regional (on land) data. Moreover, these recordings can include the near and intermediate- field terms, and hence have a greater sensitivity to magnitude. Any seafloor observatory could also include a pressure sensor capable of detecting vertical deformation on the order of 10 cm or greater. Since uplift of the seafloor in the source region is the source term for tsunami generation, this fundamental data type would greatly improve real-time tsunami forecasts for the regions immediately adjacent to the rupture zone. Owing to the greater travel time of tsunamis such data could easily be communicated to shore in time to affect warning decisions.