Ocean bottom seismometer pressure gauge observations of the 15 July 2009 Mw 7.8 Dusky Sound, New Zealand tsunami and simulations

Tsunami observations made by pressure gauges installed on Ocean Bottom Seismometers (OBSPG) have provided new opportunities in tsunami research. OBSPG observations have two main advantages over Deep-Ocean Assessment and Reporting of Tsunami (DART) records namely: 1) they come with large numbers (several tens) and dense distribution with spacing of 10-50 km versus 200-4000 km of DARTs, 2) they have high frequency with sampling rates of 40-100 Hz versus that of 0.016 Hz for DARTs. Here, we analyzed the OBSPG records of the 15 July 2009 Mw 7.8 Dusky Sound (New Zealand) tsunami. At the time of the tsunami, 30 OBSPGs equipped with differential pressure gauges (DPG) were deployed at offshore New Zealand. The tsunami also was recorded on two DARTs and four tide gauge stations. While tsunami signals were fully hidden in high-frequency recordings of the OBSPGs, we were able to clearly extract the tsunami signals by applying re-sampling, filtering, and de-convolving the DPG instrument response. In our processed OBSPG tsunami data, the tsunami arrival times were clear and the signals had periods in the range of 10-20 min which is the expected period range for a tsunami from Mw 7.8 earthquake. Numerical modeling of tsunami was conducted by using the tsunami source proposed by Beavan et al. (2010) [Geophys. J. Int. 183]. Simulations were able to fairly reproduce the observations from OBSPG, DART and tide gauge stations. However, the match for DART and tide gauge records was better than that for OBSPGs. While the observed arrival times of the first peak matched well with those of simulations for the OBSPG waveforms, the match for amplitude was not good enough. The OBS tsunami records used in this study are freely available at http://www.iris.edu.