Tsunami Source of the 2010 Mentawai, Indonesia Earthquake Inferred from Tsunami Field Survey and Waveform Modeling

The tsunami generated by the 2016 Mw 6.9 off‐Fukushima earthquake, Japan, was recorded by offshore pressure gauges on a wide, dense ocean‐bottom cable network, called S‐net, as well as by offshore GPS buoys, coastal wave gauges and coastal tide gauges. In this work, source of the tsunami was inverted from the offshore tsunami waveforms and coastal tsunami waveforms independently, using the Green's functions based on linear long‐wave theory. We found that inversion of the offshore waveforms returned robust results, more accurate and better resolved than those obtained from the coastal waveforms. Furthermore, nonlinear long‐wave simulation using the offshore‐data‐inverted tsunami source, accurately reproduced the leading waves recorded at offshore and coastal stations. These results also demonstrated that tsunami nonlinearities are non‐negligible, which is the main reason, together with weak constraints from the coastal waveforms, for the relative inaccuracy of the results of inversion from the coastal waveforms. This inaccuracy can be reduced by rearranging the conditions of inversion. Our analysis demonstrated that offshore tsunami waveforms from a wide, dense network are beneficial to deeply evaluate and improve inversion of coastal tsunami waveforms.

We explored nonlinear effects within the context of tsunami waveform inversion, wherein Green's functions were linearly superimposed to estimate earthquake slips. We focused on these effects while developing a source model for the 2003 Tokachi–Oki earthquake off Hokkaido, Japan. A source model for this earthquake was developed based on linear tsunami waveform inversion using Green’s functions and tsunami waveforms observed at tide gauge stations. Subsequently, tsunami waveforms from the source were simulated at the stations using nonlinear long-wave theory and compared with those estimated by inversion. The comparisons demonstrated that the waveforms had a non-negligible discrepancy that was attributed to advection effects, even for the primary wave used in the inversion at the two stations. This result strongly suggests that advection effects should be considered in the source modeling of the 2003 earthquake based on tsunami waveforms observed by tide gauges. Based on these results, a new tsunami waveform inversion technique that incorporates linearly approximated advection effects and maintain the framework of linear tsunami waveform inversion using Green’s functions is proposed and applied. The proposed method successfully mimicked the advection effects during the 2003 tsunami, reproduced better tsunami waveforms, and developed a source model for the 2003 earthquake using these effects. The peak slip amount and seismic moment were greater in the source model with advection effects than those without the effects. This finding suggests that the values in the source models developed for other earthquake events without considering these effects may have been underestimated.Graphical abstract

We propose a method of tsunami waveform inversion to accurately estimate a tsunami source by incorporating the effect of permanent seafloor deformation recorded by ocean‐bottom pressure gauges (OBPGs) within the source region. We developed a general expression of water‐depth fluctuation recorded at an OBPG following seafloor deformation of arbitrary spatiotemporal distribution. By assuming that coseismic rupture propagates with infinite velocity, the general expression can be reduced to an equation relating observed OBPG waveforms to initial sea‐surface displacement at the source by using a Green's function consisting of two terms: the Green's function used in regular tsunami inversion and a correction term to account for water‐depth change in response to permanent seafloor deformation. By using the two‐term Green's functions, the effect of seafloor deformation can be taken into account in tsunami source estimation. We applied the revised inversion method to observations of coseismic seafloor deformation and tsunami during the 2003 Tokachi‐oki earthquake (Mw 8.3) at two OBPG stations near the Kuril Trench. The tsunami source model we estimated is consistent with models previously derived using various other geophysical data sets. Furthermore, the coastal tsunami waveforms we modeled match the observed tsunami well. Forecasts of tsunami arrival times and first peak amplitudes by our method can be obtained 20 min after an earthquake, and can be provided to the coastal communities nearest to the source with a lead time of ∼10 min.

The source of the West Java tsunami of July 17, 2006, which was generated during a large earthquake near the Sunda trench, is constrained by tsunami waveforms that were recorded on six tide gauges around the Indian Ocean. The tsunami travel times poorly constrain the source area, probably because shallow bathymetry near these gauges is not well known. Inversion of tsunami waveforms, however, reveals that the tsunami source was about 200 km long. The largest slip, about 2.5 m for instantaneous rupture model, was located about 150 km east of the epicenter. Most of the slip occurred on shallow parts of the fault, indicating that this earthquake shares the same characteristics with “tsunami earthquakes” which generate abnormally large tsunamis compared with ground shaking. The slip distribution yields a total seismic moment of 7.0 × 1020 Nm (Mw = 7.8).

Tsunamigenic Earthquakes and Their Consequences

PreviousNext No AccessProceedings of the 11th SEGJ International Symposium, Yokohama, Japan, 18-21 November 2013Evaluation of Difference in Tsunami Wave Pressure among Different Tsunami Source ModelsAuthors: Satoru FujiharaTakahiro TamiyaMariko KorenagaNorihiko HashimotoSatoru FujiharaITOCHU Techno-Solutions Corporation, Search for more papers by this author, Takahiro TamiyaITOCHU Techno-Solutions Corporation, Search for more papers by this author, Mariko KorenagaITOCHU Techno-Solutions Corporation, Search for more papers by this author, and Norihiko HashimotoITOCHU Techno-Solutions Corporation, Search for more papers by this authorhttps://doi.org/10.1190/segj112013-137 SectionsAboutPDF/ePub ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InRedditEmail Abstract We performed tsunami simulation of 2011 Tohoku-Oki Earthquake around Fukushima I Nuclear Power Plant and Fukushima II Nuclear Power Plant in Fukushima Prefecture of Japan by using several tsunami source models in order to evaluate relative differences in things such as tsunami wave pressure. As the result, a fair amount of relative differences in maximum wave height and wave pressure for incoming tsunami onto inland region was observed. Keywords: 2011 Tohoku-Oki Earthquake, Tsunami wave force, Tsunami waveform inversion, Soma Port, Fukushima I Nuclear Power Plant, Fukushima II Nuclear Power Plant, earthquake, shallow, waterPermalink: https://doi.org/10.1190/segj112013-137FiguresReferencesRelatedDetails Proceedings of the 11th SEGJ International Symposium, Yokohama, Japan, 18-21 November 2013ISSN (online):2159-6832Copyright: 2013 Pages: 580 publication data© 2013 Published in electronic format with permission by the Society of Exploration Geophysicists of JapanPublisher:Society of Exploration Geophysicists HistoryPublished: 15 Jan 2014 CITATION INFORMATION Satoru Fujihara, Takahiro Tamiya, Mariko Korenaga, and Norihiko Hashimoto, (2013), "Evaluation of Difference in Tsunami Wave Pressure among Different Tsunami Source Models," SEG Global Meeting Abstracts : 547-550. https://doi.org/10.1190/segj112013-137 Plain-Language Summary Keywords2011 Tohoku-Oki EarthquakeTsunami wave forceTsunami waveform inversionSoma PortFukushima I Nuclear Power PlantFukushima II Nuclear Power PlantearthquakeshallowwaterPDF DownloadLoading ...

The slip distribution and seismic moment of the 2010 and 1960 Chilean earthquakes were estimated from tsunami and coastal geodetic data. These two earthquakes generated transoceanic tsunamis, and the waveforms were recorded around the Pacific Ocean. In addition, coseismic coastal uplift and subsidence were measured around the source areas. For the 27 February 2010 Maule earthquake, inversion of the tsunami waveforms recorded at nearby coastal tide gauge and Deep Ocean Assessment and Reporting of Tsunamis (DART) stations combined with coastal geodetic data suggest two asperities: a northern one beneath the coast of Constitucion and a southern one around the Arauco Peninsula. The total fault length is approximately 400 km with seismic moment of 1.7 × 1022 Nm (Mw 8.8). The offshore DART tsunami waveforms require fault slips beneath the coasts, but the exact locations are better estimated by coastal geodetic data. The 22 May 1960 earthquake produced very large, ~30 m, slip off Valdivia. Joint inversion of tsunami waveforms, at tide gauge stations in South America, with coastal geodetic and leveling data shows total fault length of ~800 km and seismic moment of 7.2 × 1022 Nm (Mw 9.2). The seismic moment estimated from tsunami or joint inversion is similar to previous estimates from geodetic data, but much smaller than the results from seismic data analysis.

Asbtract We performed tsunami waveform inversions for the Bengkulu, southern Sumatra, earthquake on September 12, 2007 (M w 8.4 by USGS). The tsunami was recorded at many tide gauge stations around the Indian Ocean and by a DART system in the deep ocean. The observed tsunami records indicate that the amplitudes were less than several tens of centimeters at most stations, around 1 m at Padang, the nearest station to the source, and a few centimeters at the DART station. For the tsunami waveform inversions, we adopted 20-, 15- and 10-subfault models. The tsunami waveforms computed from the estimated slip distributions explain the observed waveforms at most stations, regardless of the subfault model. We found that large slips were consistently estimated at the deeper part (>24 km) of the fault plane, located more than 100 km from the trench axis. The largest slips of 6–9 m were located about 100–200 km northwest of the epicenter. The deep slips may have contributed to the relatively small tsunami for its earthquake size. The total seismic moment is calculated as 4.7 × 1021 N m (M w = 8.4) for the 10-subfault model, our preferred model from a comparison of tsunami waveforms at Cocos and the DART station.

Coseismic slip distribution on the fault plane of the 1946 Nankai earthquake (Mw 8.3) was estimated from inversion of tsunami waveforms. The following three improvements from the previous study (Satake, 1993) were made. (1) Larger number of smaller subfaults is used; (2) the subfaults fit better to the slab geometry; and (3) more detailed bathymetry data are used. The inversion result shows that the agreement between observed and synthetic waveforms is greatly improved from the previous study. In the western half of the source region off Shikoku, a large slip of about 6 m occurred near the down-dip end of the locked zone. The slip on the up-dip or shallow part was very small, indicating a weak seismic coupling in that region. In the eastern half of the source region off Kii peninsula, a large slip of about 3 m extended over the entire locked zone. Large slips on the splay faults in the upper plate estimated from geodetic data (Sagiya and Thatcher, 1999) were not required to explain the tsunami waveforms, suggesting that the large slips were aseismic. Two slip distributions on the down-dip end of the plate interface, one from geodetic data and the other from tsunami waveforms, agree well except for slip beneath Cape Muroto in Shikoku. This suggests that aseismic slip also occurred on the plate interface beneath Cape Muroto.

In the subduction zone off the west coast of central Sumatra, two great earthquakes, the 2007 great Bengkulu earthquake (Mw 8.4) and the 2010 Mentawai tsunami earthquake (Mw 7.8), occurred along the plate interface. Although the moment magnitude of the 2010 earthquake was much smaller than that of the 2007 earthquake, the tsunami heights resulting from the former 2010 earthquake were higher than those resulting from the latter 2007 earthquake, indicating that tsunami heights are difficult to forecast. An advanced method for determining appropriate source models that can explain the tsunami heights along coastal areas is needed for tsunami warning purposes. In this study, fault parameters were estimated from the W-phase inversion, and fault length and width were calculated from suitable scaling relations between those and the magnitude for the 2007 and 2010 earthquakes. Tsunami numerical simulations were conducted using various slip amounts or corresponding rigidities. The best slip amount or corresponding rigidity was selected by comparing the measured and computed tsunami heights. For the 2007 Bengkulu earthquake, the measured tsunami heights are well explained using a rigidity of 3.0 × 1010 Nm−2 (7.59-m slip amount). For the 2010 Mentawai tsunami earthquake, the measured tsunami heights are well explained using a rigidity of 1.5 × 1010 Nm−2 (8.17-m slip amount). From those results, we determined the depth-dependent rigidity relation for Central Sumatra to estimate appropriate source models in our tsunami height forecasting method.

A tsunami earthquake is defined as a shock which generates extensive tsunamis but relatively weak seismic waves. A comparative study is made for the two recent tsunami earthquakes, and a subduction mechanism near a deep‐sea trench is discussed. These two earthquakes occurred at extremely shallow depths far off the coasts of the Kurile Islands and of eastern Hokkaido on October 20, 1963, and on June 10, 1975, respectively. Both can be regarded as an aftershock of the preceding larger events. Their tsunami heights and seismic wave amplitudes are compared with those of the preceding events. The results show that the time constants involved in the tsunami earthquakes are relatively long but not long enough to explain the observed disproportionality between the tsunamis and the seismic waves. The process times are estimated to be less than 100 s. The spatio‐temporal characteristics of the two events suggest that they represent a seaward and upward extension of the rupture associated with a great earthquake which did not break the free surface at the coseismic stage. The amplitude and phase spectra of long‐period surface waves and the long‐period P waveforms indicate that this extension of the rupture did not take place entirely along the lithospheric interface emerging as a trench axis. It rather branched upward from the interface in a complex way through the wedge portion at the leading edge of the continental lithosphere. This wedge portion consists in large part of thick deformable sediments. A large vertical deformation and hence extensive tsunamis result from such a branching process. A shallowest source depth, steepening of rupture surfaces, and a deformable nature of the source region all enhance generation of tsunamis. The wedge portion ruptured by a tsunami earthquake is usually characterized by a very low seismic activity which is presumably due to ductility of the sediments. We suggest that this portion fractures in a brittle way to generate a tsunami earthquake when it is loaded suddenly by the occurrence of a great earthquake and that otherwise it yields slowly. Upward branching of the rupture from the lithospheric interface produces permanent deformation of the free surface which is relative uplift landward and relative subsidence trenchward of the zone of surface break. This surface break zone geomorphologically corresponds to the lower continental slope between the deep‐sea terrace and the trench. Such a mode of permanent deformation seems to be consistent with a rising feature of the outer ridge of the deep‐sea terrace and a depressional feature of the trench. This consistency implies a causal relationship between great earthquake activities and geomorphological features near the trench.

We applied a new method to compute tsunami Green's functions for slip inversion of the 1 April 2014 Iquique earthquake using both near‐field and far‐field tsunami waveforms. Inclusion of the effects of the elastic loading of seafloor, compressibility of seawater, and the geopotential variation in the computed Green's functions reproduced the tsunami traveltime delay relative to long‐wave simulation and allowed us to use far‐field records in tsunami waveform inversion. Multiple time window inversion was applied to tsunami waveforms iteratively until the result resembles the stable moment rate function from teleseismic inversion. We also used GPS data for a joint inversion of tsunami waveforms and coseismic crustal deformation. The major slip region with a size of 100 km × 40 km is located downdip the epicenter at depth ~28 km, regardless of assumed rupture velocities. The total seismic moment estimated from the slip distribution is 1.24 × 1021 N m (Mw 8.0).

Geologic evidence has shown that unusual tsunami deposits are traced as high as 18 m above the current sea level or as far as 1–4 km inland from the shoreline on the Pacific coast of eastern Hokkaido, and that such unusual tsunamis have recurred at about 500 year interval with the most recent event in the 17th century. We computed coastal tsunami heights along the Hokkaido and Sanriku coasts and inundation at five coastal marshes in Hokkaido where the tsunami deposits were mapped. Three types of faults were tested: giant fault, tsunami earthquake and interplate earthquake models. The giant fault model, with the largest seismic moment, yields the lowest tsunami heights and smaller inundation than the distribution of tsunami deposits in Hokkaido, while the tsunami heights are largest in Sanriku. The tsunami earthquake model yields little inundation in Hokkaido and the smallest heights in Sanriku. The interplate earthquake model produces the largest tsunami heights and inundation in Hokkaido, reproducing the distribution of tsunami deposits on the Nemuro coast. The multi-segment interplate earthquake with variable slip (10 m on Tokachi and 5 m on Nemuro segment) can reproduce the distribution of tsunami deposits on the Tokachi coast as well, and considered as the best source model for the 17th century tsunami, although the Sanriku tsunami heights are more than 3 m, exceeding an inferred detection threshold of historical documents. The seismic moment is estimated as 8 × 1021N m (Mw8.5). Comparison with the recent 2003 Tokachi-oki earthquake indicates that the 17th century tsunami source was longer and located further offshore at shallower depth.

The November 20, 1960, Peru, October 20, 1963, Kurile and June 10, 1975, Kurile earthquakes are classified as tsunami earthquakes based on anomalously large tsunami excitation relative to earthquake magnitude. Long‐period surface wave analysis indicates double‐couple (faulting) mechanisms for all three events rather than single‐force mechanisms indicative of submarine landslides. The earthquakes have shallow depths (< 15 km) and are located near the trench axis and seaward of most other thrust zone events beneath the accretionary prism. Body waveform inversion indicates very shallowly dipping thrust faulting mechanisms for the three events, with dip angles of 6°–8°. Surface wave spectral amplitudes and deconvolution of SH waveforms suggests anomalously long source durations and large seismic moments relative to MS. Specifically, the 1963 Kurile event (MS 7.2) shows a duration of 85 s and a moment of 6.0 × 1027 dyn cm (MW 7.8), the 1975 Kurile event (MS 7.0) shows a duration of 60 s and a moment of 2.0 × 1027 dyn cm (MW 7.5), and the 1960 Peru event (MS 6.75) shows a time function consisting of four subevents with a total duration of 110–130 s and a seismic moment of 3.4 × 1027 dyn cm (MW 7.6). Estimated rupture velocities are about 1 km/s or less, but there is no evidence of unusually low stress drops. The August 1, 1968, Philippines event, previously classified as a tsunami earthquake, shows none of the anomalous source properties, and teleseismic tsunami height measurements are sparse; we do not consider this event a tsunami earthquake. Most of the “anomalous” tsunami excitation results from underestimation of earthquake size by MS due to the long source duration; the tsunami heights are not significantly anomalous relative to seismic moment. The slow nature of these events may result from rupture through the sedimentary rock along the basal decollement of the accretionary prism. Standard scaling laws when adjusted for the slow seismic velocity in the source region show an MW ‐ MS relationship similar to that observed for the tsunami earthquakes and predict MS saturation at about 7.3 rather than 8.0 for typical events.

This paper considers the importance of model parameterization, including dispersion, source kinematics, and source discretization, in tsunami source inversion. We implement single and multiple time window methods for dispersive and nondispersive wave propagation to estimate source models for the tsunami generated by the 2011 Tohoku‐Oki earthquake. Our source model is described by sea surface displacement instead of fault slip, since sea surface displacement accounts for various tsunami generation mechanisms in addition to fault slip. The results show that tsunami source models can strongly depend on such model choices, particularly when high‐quality, open‐ocean tsunami waveform data are available. We carry out several synthetic inversion tests to validate the method and assess the impact of parameterization including dispersion and variable rupture velocity in data predictions on the inversion results. Although each of these effects has been considered separately in previous studies, we show that it is important to consider them together in order to obtain more meaningful inversion results. Our results suggest that the discretization of the source, the use of dispersive waves, and accounting for source kinematics are all important factors in tsunami source inversion of large events such as the Tohoku‐Oki earthquake, particularly when an extensive set of high‐quality tsunami waveform recordings are available. For the Tohoku event, a dispersive model with variable rupture velocity results in a profound improvement in waveform fits that justify the higher source complexity and provide a more realistic source model.