Recordings of Long-Period Fluctuations Associated with the Passage of Three Distinct Tsunamis at Broadband Seismometers Made at the International Monitoring System (IMS) Hydroacoustic T-station H06 (Socorro Island, Mexico)

In the wake of the very large Indian Ocean tsunami on 26 December 2004, and several subsequent events in the Pacific Ocean, scientific and disaster mitigation studies have been initiated to better understand and observe tsunami propagation. This paper describes observations made on nearshore longperiod seismometers of a little known and, to our knowledge, recently discovered phenomenon associated with the passage of tsunamis near a coast. It provides further insight into this phenomenon. Observations of long-period effects of the passage of tsunamis on the horizontal components of nearshore seismometers were initially made by Yuan et al. (2005). A quantitative assessment was first made by Okal (2007). These observations were made on the Indian Ocean tsunami of December 2004. We observe similar long-period oscillations for the tsunamigenic Maule, Chile, earthquake of 27 February 2010 at two broadband seismometers used as hydroacoustic T-station of the International Monitoring System (IMS) network and located on Socorro Island, Mexico. Polarization and amplitude analysis of the long-period arrivals on the two closely spaced seismometers allows us to propose an explanation for the polarization of observations that were not explained in Okal (2007). Our preferred interpretation of the phenomenon is that the surface of the island close to the shore is tilted by the loading from the local sea elevation change due to the passage of the long-wavelength tsunami wave. Whereas Okal ignores the presence of the island in his physical analysis, this tilting of horizontal seismic sensors induced by this slow dynamic process explains observations from several tsunamis and we would expect similar observations on near-coastal continental stations. This hypothesis is confirmed by the quantitative assessment and the analysis of the data at the two broadband seismometers during the passage at Socorro of other tsunamis from the 29 September 2009 Samoa earthquake, and from the catastrophic Japanese earthquake on 11 March 2011. While examining seismic and hydroacoustic waves generated by the 27 February 2010 Maule, Chile, earthquake at IMS T-station H06, we observed long-period waves accompanying the passage of the tsunami generated by the earthquake. These long-period waves were observed principally on horizontal components. Similar observations at near-coastal seismometers had been made by Yuan et al. (2005), Hanson and Bowman (2005), and Okal (2007), for the 24 December 2004, Sumatra event, including the observation that the polarization of these waves is perpendicular to the coastline closest to the seismometers. To our knowledge, the most complete interpretation of these fluctuations is presented in Okal (2007). The author interprets the amplitude and period of the fluctuations, depending on the distance to the tsunamigenic source and the proximity of stations to the shoreline. Okal presents a statistical analysis of seismic recordings of low-frequency fluctuations of several tsunamis on coastal three component (3C) stations. Using these various observations, the author proposes that the instruments are responding to the combination of horizontal displacement, tilt, and perturbation in gravity described by Gilbert (1980). They are induced by the passage of the tsunami wave over the ocean basin. In this approximation, Okal simply ignored the island or continent structure, and assumed that the seismometer functions as an ocean-bottom instrument. According to this model, the direction of ground particle motions should coincide with the main direction of propagation of the tsunami waves. Yuan et al. (2005) also observe that the direction of polarization is toward the closest shore and explain the horizontal fluctuations via two possible mechanisms: passage of the gravitational tsunami wave recorded at the seismometers, and/or tilt of the coastal area of the island or continent due to a local loading by the passage of the gravity wave. The authors did not make a quantitative assessment of recorded low-frequency