Search

The major (Mw 8.2) intraplate normal-fault earthquake of 8 September 2017 in the Gulf of Tehuantepec (Chiapas, Mexico) generated a strong tsunami that severely impacted the nearby coasts of Mexico and Central America. Tsunami waves in the near-field area were measured by seventeen high-resolution coastal tide gauges and by three open-ocean DART stations anchored offshore from the affected region. Data from these sites, together with those from four distant DARTs, were used for comprehensive analyses of the 2017 event. De-tided sea level time series were examined to determine the statistical and spectral characteristics of the 2017 tsunami waves along the Mexican and Central American coastline. The characteristics of the recorded waves from this near-field event were compared with those from two great far-field events: the 2010 Chile and the 2011 Tohoku tsunamis. Maximum trough-to-crest wave heights for the 2017 tsunami were recorded at Puerto Chiapas (351 cm), Salina Cruz (209 cm), Acapulco (160 cm), Huatulco (137 cm) and Acajutla, El Salvador (118 cm). While maximum 2010 and 2011 tsunami waves were observed at specific “hot spots” (sites with a high Q-factor and pronounced resonant properties, such as Manzanillo and Acapulco), the “strengths” of the recorded 2017 tsunami waves were mostly determined by distance from the source. Contrary to the maximum wave heights, the general spectral properties of the tsunami signals for all three events were highly similar at a given coastal site and mainly resemble the spectral structure of background oscillations at the same site. This similarity indicates that the frequency properties of the tsunami waveforms for a steady-state tsunami signal are mainly determined by local topographic features rather than by the source parameters. Estimates of the “colour” of an event (i.e., the open-ocean tsunami frequency content) show that the 2017 Chiapas tsunami was mostly “reddish” (long-period), with 68% (DART 43413) to 87% (DART 43412) of the total tsunami energy related to waves with periods > 35 min. In contrast, the 2010 and 2011 tsunamis were “reddish-blue”, with 48–57% associated with long-period waves (> 35 min) and 52–43% with short-period waves (2–35 min). The dominant periods of the tsunami waves were mostly linked to the shape, length, and width of the source region: the larger the source and the shallower its depth, the longer the periods of the generated tsunami waves. The complicated structure of the source explains the saturated and wide frequency-band character of the tsunami spectra. Our analysis also reveals an anisotropic nature to the 2017 tsunami waves; waves that propagated northeastward along the mainland coast of North America and southeastward along the Central American coast were significantly different from those that propagated southwestward, normal to the source orientation. This aspect of the wave field appears to be related to two distinct types of waves; “trapped (edge) waves” retained on the shelf (which plays the role of a “wave guide”), and “leaky waves” that radiate into the open ocean.

The major (Mw 7.9) earthquake that struck the Gulf of Alaska near Kodiak Island on 23 January 2018 was a rare, mid-plate strike-slip event that triggered a minor trans-Pacific tsunami. An analysis of the simultaneous measurements of tsunami waveforms at 21 open-ocean sites (including three independent arrays of stations) and 27 coastal tide gauges in the Gulf of Alaska and along the coast of North America has enabled us to examine properties of the 2018 tsunami, its transformation over the continental slope and shelf, and its amplification as the waves approached the coast. Results show that the tsunami wave variance decreased monotonically along the west coast from northern British Columbia to southern Oregon. Based on the variance structure, the mean amplification factor for Tofino on the west coast of Vancouver Island (a “beacon” site with a long time series), was $$A_{RMS}^{Tof}$$ = 5.3, in good agreement with corresponding estimates for four major past events; 4.5 (2009 Samoa), 4.3 (2010 Chile), 6.3 (2011 Tohoku) and 5.2 (2012 Haida Gwaii). This variance-derived amplification for Tofino was greater than the amplification factor based on the amplitude ratio ($$A_{{}}^{Tof}$$ = 3.2). Spectral analysis of the records showed that the tsunami had a relatively large high-frequency content (i.e., was “blueish”), with nearly 90% of the total energy in the open ocean at frequencies > 1.7 cph (periods

The Tohoku (East Japan) earthquake of 11 March 2011 (M w 9.0) generated a great trans-oceanic tsunami that spread throughout the Pacific Ocean, where it was measured by numerous coastal tide gauges and open-ocean DART (Deep-ocean Assessment and Reporting of Tsunamis) stations. Statistical and spectral analyses of the tsunami waves recorded along the Pacific coast of Mexico have enabled us to estimate the principal parameters of the waves along the coast and to compare statistical features of the tsunami with other tsunamis recorded on this coast. We identify coastal “hot spots”—Manzanillo, Zihuatanejo, Acapulco, and Ensenada—corresponding to sites having highest tsunami hazard potential, where wave heights during the 2011 event exceeded 1.5–2 m and tsunami-induced currents were strong enough to close port operations. Based on a joint spectral analysis of the tsunamis and background noise, we reconstructed the spectra of tsunami waves in the deep ocean and found that, with the exception of the high-frequency spectral band (>5 cph), the spectra are in close agreement with the “true” tsunami spectra determined from DART bottom pressure records. The departure of the high-frequency spectra in the coastal region from the deep-sea spectra is shown to be related to background infragravity waves generated in the coastal zone. The total energy and frequency content of the Tohoku tsunami is compared with the corresponding results for the 2010 Chilean tsunami. Our findings show that the integral open-ocean tsunami energy, I 0, was ~2.30 cm2, or approximately 1.7 times larger than for the 2010 event. Comparison of this parameter with the mean coastal tsunami variance (451 cm2) indicates that tsunami waves propagating onshore from the open ocean amplified by 14 times; the same was observed for the 2010 tsunami. The “tsunami colour” (frequency content) for the 2011 Tohoku tsunami was “red”, with about 65% of the total energy associated with low-frequency waves at frequencies 35 min). The “red colour” (i.e., the prevalence of low-frequency waves) in the 2011 Tohoku, as well as in the 2010 Chile tsunamis, is explained by the large extension of the source areas. In contrast, the 2014 and 2015 Chilean earthquakes had much smaller source areas and, consequently, induced “bluish” (high-frequency) tsunamis.

The three great earthquakes off the coast of Chile on 27 February 2010 (Maule, Mw 8.8), 1 April 2014 (Iquique, Mw 8.2) and 16 September 2015 (Illapel, Mw 8.3) generated major transoceanic tsunamis that spread throughout the Pacific Ocean and were measured by numerous coastal tide gauges and open-ocean DART stations. Statistical and spectral analyses of the tsunami waves from the events recorded on the Pacific coast of Mexico enabled us to estimate parameters of the waves along the coast and to compare statistical features of the events. We also identified three coastal “hot spots” (sites having maximum tsunami risk): Puerto Angel, Puerto Madero and Manzanillo. Based on the joint spectral analyses of the tsunamis and background noise, we have developed a method for using coastal observations to determine the underlying spectrum of tsunami waves in the deep ocean. The “reconstructed” open-ocean tsunami spectra are in close agreement with the actual tsunami spectra evaluated from direct analysis of the DART records offshore of Mexico. We have further used the spectral estimates to parameterize the energy of the three Chilean tsunamis based on the total open-ocean tsunami energy and frequency content of the individual events.

Time series observations from instruments sited in the deep ocean and along coastal margins of the Pacific during major tsunami events in the years since 2010 were systematically processed. Examination of these records during four events, 2010 Chile (Maule), 2011 East Japan (Tohoku), 2012 Haida Gwaii and 2014 Chile (Iquique), show the prevalence of a small negative phase leading the first major positive tsunami wave, a phase that is not typically reproduced by current modelling approaches. We present leading negative phase signatures in examples from the more than 40 deep-ocean bottom pressure and approximately 200 tide gauge records investigated for this study. High sampling rate time series (15-s) were given greater weight in our investigation than the more readily available 1-min series. Careful investigation of tsunami arrival at each deep-ocean site highlights the role filtering techniques may play in misleading researchers or masking specific tsunami features such as the leading negative phase that is the basis of this study. The main focus of this investigation is to characterise the scale and repeatability of the phenomenon in support of recent similar findings rather than to provide a definitive explanation as to the cause. In general, our findings are in good agreement with and support the theoretical results of Watada et al. (J Geophys Res Solid Earth 119:4287–4310, 2014).

Deep-ocean tsunami measurements play a major role in understanding the physics of tsunami wave generation and propagation, and in improving the effectiveness of tsunami warning systems. This paper provides an overview of the history of tsunami recording in the open ocean from the earliest days, approximately 50 years ago, to the present day. Modern tsunami monitoring systems such as the self-contained Deep-ocean Assessment and Reporting of Tsunamis and innovative cabled sensing networks, including, but not limited to, the Japanese bottom cable projects and the NEPTUNE-Canada geophysical bottom observatory, are highlighted. The specific peculiarities of seafloor longwave observations in the deep ocean are discussed and compared with observations recorded in coastal regions. Tsunami detection in bottom pressure observations is exemplified through analysis of distant (22,000 km from the source) records of the 2004 Sumatra tsunami in the northeastern Pacific.

With this volume of the Pure and Applied Geophysics (PAGEOPH) topical issue “Tsunamis in the Pacific Ocean: 2011–2012”, we are pleased to present 21 new papers discussing tsunami events occurring in this two-year span. Owing to the profound impact resulting from the unique crossover of a natural and nuclear disaster, research into the 11 March 2011 Tohoku, Japan earthquake and tsunami continues; here we present 12 papers related to this event. Three papers report on detailed field survey results and updated analyses of the wave dynamics based on these surveys. Two papers explore the effects of the Tohoku tsunami on the coast of Russia. Three papers discuss the tsunami source mechanism, and four papers deal with tsunami hydrodynamics in the far field or over the wider Pacific basin. In addition, a series of five papers presents studies of four new tsunami and earthquake events occurring over this time period. This includes tsunamis in El Salvador, the Philippines, Japan and the west coast of British Columbia, Canada. Finally, we present four new papers on tsunami science, including discussions on tsunami event duration, tsunami wave amplitude, tsunami energy and tsunami recurrence.

An underwater landslide located in the central partof the eastern slope of Sakhalin Island was found andmapped during investigations under the auspices ofthe Korean–Russian Project Sakhalin Slope GasHydrates (SSGH) [1, 2]. The study region was confined to a depression oriented in the northeasterlydirection (Fig. 1a). The research in this region [1–4]demonstrated that closed depressions with round andslightly elongated forms in the northeasterly directionare characteristic peculiarities of the morphology ofthis depression (Fig. 1b). The slope angles of theirwalls are 7–10°, while the sizes are in the rangebetween 600 m and 10 km. They are located at depthsfrom 20 to 150 m. One of the largest depressions has anotable morphology. Its northern edge is quite flat(~7°), while the southern edge is steep (25–30°). Itconsists of one or a few cliffs with a total height up to100 m (Fig. 2).Two types of the sedimentary sections were foundwithin the study region. The first one is common. It ischaracterized by numerous reflecting levels that provide evidence about the layered structure of the sedimentary column. The second type of seismic section isobserved only in the southernmost depression crossedby seismic profile LV5603 (Fig. 2). Here, we distinguish a sedimentary body whose cover has a hillystructure, while the foot corresponding to a highamplitude reflecting level cuts off the underlyingreflectors. The body is acoustically transparentbecause there are no reflecting layers. Slope boundaries and diffracted reflections are seen on the highresolution section of seismic profiler SES2000DS,which is caused by the chaotic internal structure of thissedimentary body.The abovementioned peculiarities of the sedimentary cover structure together with the morphology ofthe depression give grounds to think that this sedimentary formation is a landslide. A section made usingseismic profiler SES2000DS with a resolution on theorder of 10 cm [4] does not demonstrate that the roofof the landslide is covered with horizontally locatedsediments. Taking into account that the rate of sediment deposition in this region during the Holocenewas 33 cm over 1000 years [2], the age of the landslideis less than 300 years. The breakoff wall of the landslide is very steep (25–30°). In some regions it consistsof two cliffs. Its form is undulating, and its length is22 km (Fig. 1b). The area occupied by the landslide is42 km@2, and its volume is approximately 4 km

Twenty-one papers on the 2011 Tohoku, Japan tsunami are included in Volume I of the PAGEOPH topical issue “Historical and Recent Catastrophic Tsunamis in the World.” Two papers discuss seismological aspects of the event with an emphasis on tsunami generation and warning. Five papers report the impacts and effects in Japan through field surveys of tsunami heights, building damage, and tsunami deposits or analysis of satellite data. Eight papers report the tsunami effects on other Pacific coasts, including the Kuril Islands, the USA, French Polynesia, the Galapagos Islands, Australia, and New Zealand. Three papers report on analyses of the instrumental records of the 2011 Tohoku tsunami, and two more papers report their modelling efforts of the tsunami. Several of the above papers also compare the 2011 Tohoku and 2010 Chile tsunamis.

Eighteen papers on past and recent destructive tsunamis are included in Volume II of the PAGEOPH topical issue “Historical and Recent Catastrophic Tsunamis in the World.” Three papers discuss deep-sea (DART) and coastal tsunami observations, warning systems and risk management in the Pacific Ocean. Four papers examine the 1755 Lisbon, 1964 Alaska, 2003 Algeria, and 2011 Haiti tsunamis. Four more papers, as well as some papers in Volume I, report on various aspects of the 2010 Chile tsunami. Two papers present some results of field survey and modelling investigation of the 2010 Mentawai, Indonesia, tsunami. Three papers report on modelling efforts of tsunami generation by earthquake and landslide, and of tsunami propagation. Finally, two papers discuss hazard assessment using a probabilistic approach.

The major (M w = 8.8) Chilean earthquake of 27 February 2010 generated a trans-oceanic tsunami that was observed throughout the Pacific Ocean. Waves associated with this event had features similar to those of the 1960 tsunami generated in the same region by the Great (M w = 9.5) 1960 Chilean Earthquake. Both tsunamis were clearly observed on the coast of British Columbia. The 1960 tsunami was measured by 17 analog pen-and-paper tide gauges, while the 2010 tsunami was measured by 11 modern digital coastal tide gauges, four NEPTUNE-Canada bottom pressure recorders located offshore from southern Vancouver Island, and two nearby open-ocean DART stations. The 2010 records were augmented by data from seven NOAA tide gauges on the coast of Washington State. This study examines the principal characteristics of the waves from the 2010 event (height, period, duration, and arrival and travel times) and compares these properties for the west coast of Canada with corresponding properties of the 1960 tsunami. Results show that the 2010 waves were approximately 3.5 times smaller than the 1960 waves and reached the British Columbia coast 1 h earlier. The maximum 2010 wave heights were observed at Port Alberni (98.4 cm) and Winter Harbour (68.3 cm); the observed periods ranged from 12 min at Port Hardy to 110–120 min at Prince Rupert and Port Alberni and 150 min at Bamfield. The open-ocean records had maximum wave heights of 6–11 cm and typical periods of 7 and 15 min. Coastal and open-ocean tsunami records revealed persistent oscillations that “rang” for 3–4 days. Tsunami energy occupied a broad band of periods from 3 to 300 min. Estimation of the inverse celerity vectors from cross-correlation analysis of the deep-sea tsunami records shows that the tsunami waves underwent refraction as they approached the coast of Vancouver Island with the direction of the incoming waves changing from an initial direction of 340° True to a direction of 15° True for the second train of waves that arrived 7 h later after possible reflection from the Marquesas and Hawaiian islands.

Major earthquakes occurred in the region of the Central Kuril Islands on November 15, 2006 (Mw = 8.3) and January 13, 2007 (Mw = 8.1). These earthquakes generated strong tsunamis recorded throughout the entire Pacific Ocean. The first was the strongest trans-Pacific tsunami of the past 42 years (since the Alaska tsunami in 1964). The high probability of a strong earthquake (Mw ≥ 8.5) and associated destructive tsunami occurring in this region was predicted earlier. The most probable earthquake source region was investigated and possible scenarios for the tsunami generation were modeled. Investigations of the events that occurred on November 15, 2006, and January 13, 2007, enabled us to estimate the validity of the forecast and compare the parameters of the forecasted and observed earthquakes and tsunamis. In this paper, we discuss the concept of “seismic gaps,” which formed the basis for the forecast of these events, and put forward further assumptions about the expected seismic activity in the region. We investigate the efficiency of the tsunami warning services and estimate the statistical parameters for the observed tsunami waves that struck the Far Eastern coast of Russia and Northern Japan. The propagation and transformation of the 2006 and 2007 tsunamis are studied using numerical hydrodynamic modeling. The spatial characteristics of the two events are compared.

Alberni Inlet is a long and narrow fjord adjacent to Barkley Sound on the Pacific Coast of Vancouver Island, Canada. Port Alberni, at the head of the inlet, was affected in 1964 by the largest Pacific tsunami waves in Canadian history. We use observations and results from two numerical models to investigate the resonant characteristics of the region and amplification of tsunami waves in Barkley Sound and Alberni Inlet. The first model (A) was forced at its open boundary with a stationary autoregressive signal, similar to the observed background noise. The second model (B) used an initial sea-level deformation from a potential earthquake off California in the southern segment of the Cascadia Subduction Zone, producing transient tsunami waves. Spectral, cross-spectral and frequency-time (f-t) analyses of the observations were used to examine the resonant properties and topographic response of the local area. The respective results show large admittance functions over a wide 0.5–0.9 cph frequency band, implying a low Q factor but high amplification of arriving waves. This unusual behavior is a result of two effects: A quarter-wave resonance of the system for its fundamental Helmholtz mode and amplification due to the narrowing of the channel cross section from Barkley Sound to Alberni Inlet. The model A numerical results agree favorably with the observations, indicating an energetic resonant mode at frequency of ~0.53 cph (112 min), with its nodal line located near the entrance to Barkley Sound and amplification factor value close to 12. The results from the tsunami propagation model (B) yield spectral characteristics similar to those from the model A and from the observations. The maximum tsunami current speed for this scenario is 2.4 ms-1 in Sproat Narrows, which divides Alberni Inlet into two parts, while the largest computed wave height is 1.6 m in the northern Alberni Inlet, in the area of Port Alberni.

This study presents analyses and numerical simulations of local tsunamis generated by two recent earthquakes off the coast of British Columbia. The Queen Charlotte Islands earthquake (Mw = 6.1) on 12 October 2001 generated waves that were recorded by tide gauges at Bamfield, Tofino, Winter Harbour and Port Hardy on the coast of Vancouver Island, with maximum measured wave heights of 11.3, 18.2, 22.7 and 14.5 cm, respectively. The Explorer Plate earthquake off Vancouver Island (Mw = 6.6) on 2 November 2004 generated waves recorded only at Bamfield and Tofino, with lower heights of 7.5 cm and 10.8 cm, respectively. The generation of tsunamis by these moderate, Mw = 6.1–6.6, local earthquakes suggests the possibility of destructive tsunamis from local sources other than the Cascadia Subduction Zone. This, in turn, has implications for tsunami hazards for this seismically active region of coastal North America.

On November 15, 2006, at 11:14 UTC, one of the strongest tsunamigenous earthquakes occurred near the Russian coast in the region of the Central Kuril Islands. Its magnitude was M w = 8.3, and the epicenter was located on the island slope of the deep Kuril–Kamchatka Trench 90 km southeast of Simushir Island (Fig. 1). The earthquake generated a tsunami that was manifested over the entire Pacific Ocean and was recorded with high accuracy by numerous coastal pressure gauges and by a large number of deep-water stations located in the open part of the ocean. The tsunami of 2006 became the strongest event of such type observed in the Pacific Ocean over the last 42 years (after the strongest Alaska tsunami in 1964). It occurred less than two years after the catastrophic December 26, 2004, earthquake and tsunami in the Indian Ocean, which caused unprecedented destruction and victims. However, it stimulated a drastic intensification in tsunami research [1, 2]. The November 15, 2006, earthquake and tsunami are unique because the possibility of their occurrence in the Central Kuril Islands region was predicted [1, 3, 4] and the possible scenarios of tsunami propagation were modeled [5, 6].

Analysis of sea-level data obtained from the Atlantic Global Sea Level Observing System (GLOSS) sea-level station at Takoradi, Ghana, West Africa, clearly reveals a tsunami signal associated with the Mw = 9.3 Sumatra earthquake of 26 December 2004 in the Indian Ocean. The tsunami arrived at this location on 27 December 2004 at approximately 01:38 UTC (which is close to the expected tsunami arrival time at that site), after travelling for more than 24 hours. The first wave was negative (trough), in contrast with the South African stations where the first wave was mainly positive (crest). The dominant observed period at Takoradi was about 42 minutes. The maximum trough-to-crest wave height (41cm) was observed on 28 December at 00:15 UTC. There were two distinct tsunami ‘bursts', separated in time by about 14 hours, the larger being the second burst. A small residual lowering of the sea level (~15cm) during the tsunami and for several days afterwards, and a delayed (~4.5 days) lowering of seawater temperature (up to ~4.5&#176C), was observed, possibly indicating the presence of internal waves through the Gulf of Guinea associated with propagating tsunami waves. The prominent tsunami signal found in the Takoradi record suggests that tsunami waves could also be found at other sites off the West African coast. Keywords: Atlantic Ocean, focusing, Ghana, GLOSS, Gulf of Guinea, internal wave, reflection, sea-level changes, tide gauge, tsunami wave, West AfricaAfrican Journal of Marine Science 2006, 28(3&4): 705–712

Data for tsunamigenic earthquakes and observed tsunami run-up are used to estimate tsunami-risk for the coasts of Peru and northern Chile for zones bounded by 5–35° S latitude. Tsunamigenic earthquake estimates yield magnitudes of 8.52, 8.64, and 8.73 for recurrence periods of 50, 100, and 200 years, respectively. Based on three different empirical relations between earthquake magnitudes and tsunamis, we estimate expected tsunami wave heights for various return periods. The average heights were 11.2 m (50 years), 13.7 m (100 years), and 15.9 m (200 years), while the maximum height values (obtained by Iida’s method) were: 13.9, 17.3, and 20.4 m, respectively. Both the “averaged” and “maximum” seismological estimates of tsunami wave heights for this region are significantly smaller than the actually observed tsunami run-up of 24–28 m, for the major events of 1586, 1724, 1746, 1835, and 1877. Based directly on tsunami run-up data, we estimate tsunami wave heights of 13 m for a 50-year return period and 25 m for a 100-year return period. According to the “seismic gap” theory, we can expect that the next strong earthquake and tsunami will occur between 19 and 28° S in the vicinity of northern Chile.

On November 18, 1929, a M=7.2 earthquake occurred at the southern edge of the Grand Banks, 280 km south of Newfoundland. The earthquake triggered a large submarine slope failure (200 km 3 ), which was transformed into a turbidity current carrying mud and sand eastward up to 1000 km at estimated speeds of about 60–100 km/h, breaking 12 telegraph cables. The tsunami generated by this failure killed 28 people, making it the most catastrophic tsunami in Canadian history. Tsunami waves also were observed along other parts of the Atlantic coast of Canada and the United States. Waves crossing the Atlantic were recorded on the coasts of Portugal and the Azores Islands. Tsunami waves had amplitudes of 3–8 m and runup of up to 13 m along the coast of the Burin Peninsula (Newfoundland). To simulate the slide-generated tsunami from this event, our initial analysis uses a shallow-water model; the slide was assumed to be a viscous, incompressible fluid layer; water was inviscid and incompressible. The preliminary results of the numerical modeling are encouraging, with computed and observed tsunami arrival times in reasonable agreement. D 2004 Elsevier B.V. All rights reserved.

A modified and corrected version of the viscous slide model of JIANG and LEBLOND (1994) is used to assess the tsunami risk associated with hypothetical underwater slope failures in two coastal areas of British Columbia having potentially unstable sediment deposits: (a) Malaspina Strait, separating the mainland coast and Texada Island in the central Strait of Georgia; and (b) Roberts Bank on the foreslope of the Fraser River Delta in the southern Strait of Georgia. The intent of this study is to demonstrate the capability of the model for tsunami risk assessment and to improve upon previous studies of tsunami risk in the region based on reasonable submarine landslide scenarios. The potential risk from tsunamis associated with slide failures has been examined, but the likelihood of failure events themselves was not considered. For the Malaspina Strait scenarios, simulated tsunamis are generated by failure of a lobe of perched sediment situated on the slope of eastern Texada Island. Failure as a flow slide of the estimated 1,250,000 m 3 of sediment generates wave troughs reaching � 4:9 m and trough-to-crest heights of 6 to 8 m along the coast of Texada Island. At Cape Cockburn, on the opposite side of the strait, wave heights of 1.5 to 2.0 m are produced. For Roberts Bank, simulated waves are examined for two separate failure scenarios. The larger slide (Case 1) involves the failure of a sediment lobe with lateral dimensions of 7 · 3k m 2 and volume of 0.75 km 3 while the smaller slide (Case 2) fails a sediment lobe with dimensions of 4 · 2.6 km 2 and volume of 0.23 km 3 . Computations were made both for high (+3 m) and low (� 3 m) tide conditions. For both failure volumes, maximum wave amplitudes (up to 18 m for Case 1 and 8 m for Case 2) occur on the coasts of Mayne and Galiano Islands, opposite the source area. Wave amplitudes are much smaller (1 to 4 m) on the mainland coast because of the reflection of the initial waves from Roberts Bank. Additional numerical experiments were conducted for both regions to estimate the sensitivity of the computed tsunami wave heights to input parameters, such as slide viscosity, bulk density, and slide position.

In light of the recent enhanced activity in the study of tsunami waves and their source mechanisms, we consider tsunami-like waves that are induced by atmospheric processes rather than by seismic sources. These waves are mainly associated with atmospheric gravity waves, pressure jumps, frontal passages, squalls and other types of atmospheric disturbances, which normally generate barotropic ocean waves in the open ocean and amplify them near the coast through specific resonance mechanisms (Proudman, Greenspan, shelf, harbour). The main purpose of the present study is to describe this hazardous phenomenon, to show similarities and differences between seismic and meteorological tsunamis and to provide an overview of meteorological tsunamis in the World Ocean. It is shown that tsunamis and meteotsunamis have the same periods, same spatial scales, similar physical properties and affect the coast in a comparably destructive way. Some specific features of meteotsunamis make them akin to landslide-generated tsunamis. The generation efficiency of both phenomena depend on the Froude number (Fr), with resonance taking place when Fr~1.0. Meteotsunamis are much less energetic than seismic tsunamis and that is why they are always local, while seismic tsunamis can have globally destructive effects. Destructive meteotsunamis are always the result of a combination of several resonant factors; the low probability of such a combination is the main reason why major meteotsunamis are infrequent and observed only at some specific locations in the ocean.

Extreme atmosphere-induced seiche oscillations occasionally occur in specific inlets and bays of the world ocean causing severe damage to coastal areas, ships and port constructions. Ciu- tadella inlet (Menorca Island, Western Mediterranean) can be singled out as a place where such large seiches, locally known as rissaga, are quite common. Similar (although weaker) oscillations are also regularly observed in bays of Shikotan Island (South Kuril Islands, northwestern Pacific). Several spectacular events in these regions, identified in the first part of this study (Rabinovich and Monserrat, 1996), are analysed to determine the atmospheric parameters responsible for the generation of large- amplitude seiches. Their generation mechanism was shown to be quite different from that causing ordinary background oscillations. Coincidence of some external factors and certain resonance effects seem to be necessary to produce the destructive waves. In particular, rissaga waves in Ciutadella inlet were found to be related to significant atmospheric disturbances propagating from the southwest, coinciding with the orientation of the inlet, and having a phase speed of about 30 m/s, which is close to the phase speed of long waves offshore from Menorca. Pronounced resonant properties of the inner basin strongly amplify incoming waves in Ciutadella inlet. In contrast, the bays of the northwestern coast of Shikotan Island are protected from normally incident atmosphere-induced waves by the elongated Kunashir Island, hence the whole situation there is not so favorable for the excitation of large seiches.

The catastrophic Ligurian Earthquake on February 23, 1887 caused significant damage and numerous casualties in northwestern Italy and southeastern France. The first (main) shock (estimated magnitude of 6.2–6.5) produced tsunami waves observed along 250 km of the Ligurian coast from Genoa to Cannes. Typical tsunami run-up heights were 1–2 m. The first wave was found to be negative, supporting the assumption that the earthquake was produced by release of stress along an offshore normal fault. Computed tsunami waveforms are in qualitative agreement with observed run-ups. Unique 1887 tsunami records in Genoa and Nice harbors were found and analyzed. Numerical computations of eigenoscillations for the old (1887) Genoa harbor were made and found to be in a good agreement with the tsunami observations. The dominant observed period of 22.5 min was shown to be related to the fundamental (Helmholtz) mode of the old port. Long “ringing” and slow decay of tsunami oscillations in Genoa are associated with “pumping” of wave energy from the external basin. Several wave trains are revealed in the record that occur at 2–2.5 h intervals. They are interpreted as edge waves propagating along the Ligurian shelf and reflecting from its borders.

A new approach is proposed to separate the influence of source and topography in observed tsunami spectra and to reconstruct the initial source spectrum. The method assumes a linear tide gauge response to external forcing and is based on comparative analysis of tsunami and background spectra. Evaluation of the ratio of tsunami to background spectra gives functional characteristics that are invariant of station location and dependent only on the source parameters. Three events, the Urup tsunami of December 22, 1991 (northwest Pacific), the Hokkaido Southwest (Okushiri) tsunami of July 12, 1993 (Sea of Japan), and the Shikotan tsunami of October 4, 1994 (northwest Pacific), are used as examples of the proposed approach. The source spectra reconstructed from the analysis of different tide gauges were found to be in good agreement with each other and with the seismological, geodetic, and field survey data.

We show that the tsunami of November 3, 1994 in Skagway, Alaska was generated by an underwater landslide formed during the collapse of a cruise ship wharf undergoing construction at the head of Taiya Inlet. This event occurred at a time of extreme low tide and was not associated with a regional seismic event or incoming oceanic tsunami. Persistent wave motions with an amplitude of 1 m and a period of 3 min recorded by a tide gauge in Skagway Harbor following the landslide are linked to the formation of a cross-inlet seiche and quarter-wave resonance within the harbor. The high Q factor for the harbor (Q ≈ 20) indicates weak dissipation and strong resonance within the harbor.

[1] The 2004 Sumatra tsunami was clearly recorded by two UK bottom pressure gauges, DPN and DPS, deployed in Drake Passage between South America and Antarctica. These open-ocean records were examined to estimate characteristics of the tsunami waves and to compare the results of numerical simulations with the observations. Maximum wave heights measured at these gauges were 4.9 cm at DPN and 7.4 cm at DPS; the travel times from the source area were 19 h 46 min and 19 h 39 min respectively, consistent with the times obtained from the nearby coastal tide gauges. The numerical model described well the frequency content, amplitudes and general structure of the observed waves, with only small time shifts probably related to wave dispersion effects. The shifts were 15 min for DPN and 10 min for DPS, with the modeled waves leading the observations in each case. Further inspection of the simulated and observed records revealed that the identified tsunami waves are related to the second (main) train of waves propagating by the energy conserving route along the mid-ocean ridges, while the first train of waves travelling by the fastest route across the ocean remained unrecognizable in the observed DPS and DPN records and undetectable in the records of coastal tide gauges because of their insignificant amplitudes compared to the background variability.

[1] As part of the NEPTUNE-Canada cabled seafloor observatory, an array of six high-precision bottom pressure recorders was installed in the late summer of 2009 at depths from 100 to 2600 m seaward of the southwest coast of Vancouver Island in the northeast Pacific. The instruments transmit 1-sec bottom pressure data at roughly 0.1 mm equivalent sea level height to the Data Management and Archiving System (DMAS) at the University of Victoria. On September 30, 2009, the array recorded waves of 2.5 to 6 cm amplitude associated with the transoceanic tsunami generated by the Mw = 8.1 Samoa earthquake in the South Pacific. These open-ocean observations were uncontaminated by coastal effects, demonstrating that NEPTUNE records from future tsunami events can be effectively used as realtime input to a regional numerical tsunami forecast model. We validate this proposition by showing that wave forms simulated by the regional model using the leading train of waves of the 2009 event are in good agreement with observed tsunami records for both the shelf stations and nearby coastal tide gauges. Tsunami waves simulated by this model are also significantly more accurate for local regions than those determined by global-scale tsunami models. This ability to assimilate “pristine” open-ocean data from the cabled observatory into an operational tsunami forecast model makes it possible to provide updated wave information that could help mitigate the impact of future tsunamis approaching the west coast of British Columbia and northern Washington State.

[1] We report unique deep-sea recordings of the Sumatra tsunami of December 2004 by high-resolution DART® (Deep-ocean Assessment and Reporting of Tsunami) and CORK (Circulation Obviation Retrofit Kit) bottom pressure sensors deployed at depths of ∼1500–3500 m in Cascadia Basin in the northeast Pacific. The simultaneous records from these sites establish the first-ever regional-scale tsunami detection array for the open ocean, enabling us to resolve both seafloor and crustal tsunami signals and to determine fundamental properties of the waves following their 22,000 km journey from the source region. Waves reaching the basin had mean amplitudes of ∼5 mm with energy spread over a broad frequency band from 0.4 to 7 cph. Peak tsunami energy was in the 0.8 to 2 cph (75 to 30 min) band. Leading waves from the event arrived 34–35 h after the earthquake, roughly 7 h later than expected, suggesting that the tsunami mainly propagated by the “most economic” (minimum energy loss) path along mid-ocean ridge wave-guides rather than taking the direct and fastest path across the ocean. Motions within the peak energy band comprise roughly 50% coherent progressive waves propagating from the south at longwave phase speeds of ∼150 ms−1 and 50% random waves scattered from the coast and bottom irregularities. Tsunami amplitudes in the borehole were one-third those on the seafloor. The extensive “ringing” and anomalously slow (3.5-day) e-folding decay time of the tsunami wave energy indicates long duration energy flux radiating from the Indian Ocean via the Southern Pacific Ocean.

[1] Bottom pressure measurements acquired from the TAG hydrothermal field on the Mid-Atlantic Ridge (26°N) contain clusters of narrowband spectral peaks centered at periods from 22 to 53.2 minutes. The strongest signal at 53.2 min corresponds to 13 mm of water depth variation. Smaller, but statistically significant, signals were also observed at periods of 22, 26.5, 33.4, and 37.7 min (1–4 mm amplitude). These kinds of signals have not previously been observed in the ocean, and they appear to represent vertical motion of the seafloor in response to hydrothermal flow - similar in many ways to periodic terrestrial geysers. We demonstrate that displacements of 13 mm can be produced by relatively small flow-induced pressures (several kPa) if the source region is less than ∼100 m below the seafloor. We suggest that the periodic nature of the signals results from a non-linear relationship between fluid pore pressure and crustal permeability.

In the last 15 years there have been 16 tsunami events recorded at tide stations on the Pacific Coast of Canada. Eleven of these events were from distant sources covering almost all regions of the Pacific, as well as the December 26, 2004 Sumatra tsunami in the Indian Ocean. Three tsunamis were generated by local or regional earthquakes and two were meteorological tsunamis. The earliest four events, which occurred in the period 1994–1996, were recorded on analogue recorders; these tsunami records were recently re-examined, digitized and thoroughly analysed. The other 12 tsunami events were recorded using digital high-quality instruments, with 1-min sampling interval, installed on the coast of British Columbia (B.C.) in 1998. All 16 tsunami events were recorded at Tofino on the outer B.C. coast, and some of the tsunamis were recorded at eight or more stations. The tide station at Tofino has been in operation for 100 years and these recent observations add to the dataset of tsunami events compiled previously by S.O. Wigen (1983) for the period 1906–1980. For each of the tsunami records statistical analysis was carried out to determine essential tsunami characteristics for all events (arrival times, maximum amplitudes, frequencies and wave-train structure). The analysis of the records indicated that significant background noise at Langara, a key northern B.C. Tsunami Warning station located near the northern end of the Queen Charlotte Islands, creates serious problems in detecting tsunami waves. That station has now been moved to a new location with better tsunami response. The number of tsunami events observed in the past 15 years also justified re-establishing a tide gauge at Port Alberni, where large tsunami wave amplitudes were measured in March 1964. The two meteorological events are the first ever recorded on the B.C. coast. Also, there have been landslide generated tsunami events which, although not recorded on any coastal tide gauges, demonstrate, along with the recent investigation of a historical catastrophic event, the significant risk that landslide generated tsunami pose to coastal and inland regions of B.C.

A destructive tsunami-like event (locally known as ‘ rissaga’ waves) occurring on 15 June, 2006 in Ciutadella Harbour (Menorca, Balearic Islands) is reproduced by a numerical model forced by a travelling atmospheric disturbance. The disturbance is reconstructed from microbarograph measurements, being the only available instrumental data at the time of the event. The model is verified based on two weaker 1997 events, which were recorded by a number of bottom pressure recorders operating at that time on the Menorca shelf, in Ciutadella Inlet and adjacent Platja Gran Inlet. Both 1997 events are numerically simulated and good agreement is achieved with observations in time, frequency (including eigenfrequencies of the affected inlets) and wave heights. Subsequently the same model is applied to simulate the 2006 event. The vigorous currents with speeds up to 400 cm/s are found to occur specifically at those areas of the harbour where the most severe damage and sinking of boats had been reported. Maximum simulated sea-level heights of 2.5 m were about one half of those reported by eyewitnesses. This difference is apparently caused by quality and spatial resolution of bathymetry data. However, in general, the model is capable of reproducing the event fairly well and can probably be used for future assessment and mitigation activities on the coasts of the Balearic Islands.

Two major earthquakes near the Central Kuril Islands (Mw=8.3 on 15 November 2006 and Mw=8.1 on 13 January 2007) generated trans-oceanic tsunamis recorded over the entire Pacific Ocean. The strongest oscillations, exceeding several meters, occurred near the source region of the Kuril Islands. Tide gauge records for both tsunamis have been thoroughly examined and numerical models of the events have been constructed. The models of the 2006 and 2007 events include two important advancements in the simulation of seismically generated tsunamis: (a) the use of the finite failure source models by Ji (2006, 2007) which provide more detailed information than conventional models on spatial displacements in the source areas and which avoid uncertainties in source extent; and (b) the use of the three-dimensional Laplace equation to reconstruct the initial tsunami sea surface elevation (avoiding the usual shallow-water approximation). The close agreement of our simulated results with the observed tsunami waveforms at the open-ocean DART stations support the validity of this approach. Observational and model findings reveal that energy fluxes of the tsunami waves from the source areas were mainly directed southeastward toward the Hawaiian Islands, with relatively little energy propagation into the Sea of Okhotsk. A marked feature of both tsunamis was their high-frequency content, with typical wave periods ranging from 2–3 to 15–20 min. Despite certain similarities, the two tsunamis were essentially different and had opposite polarity: the leading wave of the November 2006 trans-oceanic tsunami was positive, while that for the January 2007 trans-oceanic tsunami was negative. Numerical modeling of both tsunamis indicates that, due to differences in their seismic source properties, the 2006 tsunami was more wide-spread but less focused than the 2007 tsunami.

The 2004 Sumatra tsunami propagated throughout the World Ocean and was clearly recorded by tide gauges on the Atlantic coast of South America. A total of 17 tsunami records were found and subsequently examined for this region. Tsunami wave heights and arrival times are generally consistent with numerical modeling results. Maximum wave heights of more than 1.2 m were observed on the coasts of Uruguay and southeastern Brazil. Marked differences in tsunami height from pairs of closely located tide gauge sites on the coast of Argentina illustrate the importance that local topographic resonance effects can have on the observed wave response. Findings reveal that, outside the Indian Ocean, the highest waves were recorded in the South Atlantic and not in the Pacific as has been previously suggested.

[1] A detailed analysis of over one hundred tide gauge records from the Atlantic coast of North America reveals that the arrival of the 26 December 2004 Sumatra tsunami on this coast coincided with the presence of tsunami-like waves being generated by a major storm tracking northward along the eastern seaboard of the United States. According to the tide gauge records, waves from the two events coalesced along the shores of Maine and Nova Scotia on 27 December where they produced damaging waves with heights in excess of 1 m. Tsunami waves were identified in almost all outer tide gauges from Florida to Nova Scotia with maximum tsunami heights for the northern regions estimated to be 32–39 cm. In the south, maximum tsunami wave heights were in the range of 15 to 33 cm.

Numerical model simulations, combined with tide-gauge and satellite altimetry data, reveal that wave amplitudes, directionality, and global propagation patterns of the 26 December 2004 Sumatra tsunami were primarily determined by the orientation and intensity of the offshore seismic line source and subsequently by the trapping effect of mid-ocean ridge topographic waveguides.

[1] Wave arrival times obtained from coastal tide gage and satellite altimetry records for the Indian Ocean are used to delineate the source region for the December 26, 2004 Sumatra tsunami. Findings define a curved, 250-km wide, 1000-km long tsunami source region centered over the Sunda Subduction Zone, which closely matches the seismic source estimated from broadband geophysical data. Imbedded in this general region are “hot spots” associated with the southern fast-slip and northern slow-slip domains which served as distinct source areas for the destructive waves that inundated the coast of the Indian Ocean.

Recent catastrophic tsunamis at Flores Island, Indonesia (1992), Skagway, Alaska (1994), Papua New Guinea (1998), andIzmit, Turkey (1999) have significantly increased scientific interest in landslides, and slide-generated tsunamis. Theoretical investigations and laboratory modeling further indicate that purely submarine landslides are ineffective at tsunami generation compared with subaerial slides. In the present study, we undertook several numerical experiments to examine the influence of the subaerial component of slides on surface wave generation, and to compare the tsunami generation efficiency of viscous, and rigid-body slide models. We found that a rigid-body slide produces much higher tsunami waves than a viscous (liquid) slide. The maximum wave height, and energy of generated surface waves were found to depend on various slide parameters, and factors, including slide volume, density, position, and slope angle. For a rigid-body slide, the higher the initial slide above sea level, the higher the generated waves. For a viscous slide, there is an optimal slide position (elevation) which produces the largest waves. An increase in slide volume, density, and slope angle always increases the energy of the generated waves. The added volume associated with a subaerial slide entering the water is one of the reasons that subaerial slides are much more effective tsunami generators than submarine slides. The critical parameter determining the generation of surface waves is the Froude number, Fr (the ratio between slide, and wave speeds). The most efficient generation occurs near resonance when Fr = 1.0. For purely submarine slides with p 2 ≤0.2 g-cm-3, the Froude number is always less than unity, and resonance coupling of slides, and surface waves is physically impossible. For subaerial slides there is always a resonant point (in time and space) where Fr = 1.0 for which there is a significant transfer of energy from a slide into surface waves. This resonant effect is the second reason that subaerial slides are much more important for tsunami generation than submarine slides.

A three-dimensional, shallow-water numerical model for a viscous landslide with full slide-wave interaction (Kulikov et al.; 1996; Fine et al., 1998) has been modified to include the subaerial component of the landslide. The model is used to simulate the November 3, 1994 tsunami in Skagway, Alaska generated by collapse of the PARN Dock. Results show that the dock slide moved down the steep (30–35°) slope of Taiya Inlet and was guided along the trough at the base of the slope, consistent with geomorphological findings. The leading tsunami wave, propagating in front of the advancing slide, impacted the Alaska State Ferry Terminal and the NOAA tide gauge site as a positive wave (crest), consistent with the tide gauge record and with the results of laboratory modelling by Raichlen et al. (1996). Computed wave heights for the PARN Dock failure (13 m at the Ferry Terminal, 7.7 m at the tide gauge site, and 1.3 in the Small Boat Harbor) agree closely with the tide gauge record and eyewitness accounts. The computed 3.0 min period for the fundamental long-wave mode for Skagway Harbor is nearly identical to the observed period. Estimates of the Q-factor (Q≈24) are comparable to observed values (Q≈21), suggesting significant tsunami energy retention in the harbour. Energy loss appears to be through radiation damping rather than from frictional effects. A detailed examination of the slide motion and associated tsunami waves in the vicinity of the PARN Dock reveals that, in the first few seconds, a “wall of water” would have formed opposite the dock and that the floating Ferry Terminal would have been impacted 15 to 20 s after onset of the event, consistent with eyewitness accounts. The floating debris observed at the still-standing northern portion of the dock was apparently carried alongshore by a secondary wave crest originating near the collapsed southern part of the dock.