A tsunami (plural: tsunami or tsunamis) is a series of water waves caused by an impulsive disturbance to a body of water. Tsunamis are most commonly caused in the ocean through the deformation of the seafloor during an earthquake or as a result of a submarine landslide. Additional source mechanisms include volcanic eruptions, sub-aerial landslides, asteroids falling into a body of water, or through meteorological forcing (also referred to as ‘meteo-tsunami’). The word tsunami is Japanese (Kanji: 津波) and translates literally to ‘harbor wave’; due to the fact that this phenomenon was frequently observed in bays and harbors in Japan. Tsunamis are also referred to as ‘tidal waves’ and ‘seismic sea waves’.
Tsunamis can generally be categorized as ‘near field’ or ‘far field’ depending on where the majority of its effects occur. A far-field tsunami (also ‘tele-tsunami’ or ‘trans-oceanic tsunami’) is one which causes damage or serious effects at great distances from the source region while the effects of a near field or ‘local’ tsunami are constrained roughly to the tsunami source region. This term is somewhat ambiguous in that tsunamis with appreciable effects in the far field also produce significant effects in the near-field and thus have both near and far-field components.
Tectonically generated tsunami
Most destructive tsunamis have been caused by thrust earthquakes along subdutction zones such as those existing around the Pacific Rim and the Sunda Arc south and west of Indonesia. Notable recent examples include the Pacific tsunamis of 1960 (Chile), 1964 (Alaska), 2011 (Japan) and the 2004 Indian Ocean Tsunami (north Sumatra). Another common tectonic source mechanism is that of the outer-rise earthquake characterized by normal faulting and subsequent large scale subsidence in the oceanic tectonic plate seaward of a subduction zone trench; the 1933 Sanriku tsunami and 2009 Samoa (South Pacific) tsunami are examples of this type of generation mechanism. Tsunamis have also been caused through strike slip faulting where one side of the fault is displaced laterally relative to the other and there is very little vertical deformation of the sea bed. However, these types of tsunami may also be associated with additional source mechanisms such as submarine slumps or landslides which contribute to the overall strength of the tsunami and its impact.
Amongst the subduction zone thrust-type earthquakes exists a special class of events where the resulting tsunami is much larger than would be expected based on the earthquake magnitude alone. Such events have been termed ‘tsunami earthquakes’ (Kanamori, 1972) and are generally characterized by earthquake rupture occurring on the shallower portion of the subduction zone plate interface, a slower rupture velocity and rate of seismic energy release, and a larger sea floor deformation relative to other earthquakes of similar magnitude. Each of these components contributes to an overall larger tsunami, particularly in the near field. Examples of this type of event include Sanriku Japan, 1896, Nicaragua, 1992, East Java, Indonesia, 1994, West Java Indonesia, 2005 and the Mentawai Islands, Indonesia, 2010.
Landslide generated tsunami
Tsunami can also be generated by slope failures – commonly referred to as landslides, slumps, slips or debris flows. Tsunamigenic slope failures can occur either above or below water (i.e. ‘subaerial’ or ‘submarine’). Tsunami waves are generated as water is displaced; in the case of a subaerial event, as the material crashes on to the water surface or when submarine, as water above the slide is pulled downward in to the void left as the slide material slides down slope. Tsunamigenesis by slope failure is an active and on-going area of scientific research. The key parameters controlling the height of the initial wave are the slope of the sea bed, the amount of material displaced, the thickness of the slide, the depth at which the slide occurs, the speed at which it fails and the distance over which the slide material translates (Bardet et al., 2003). Also critical to the study of landslide generated tsunami is the ability to define the recurrence intervals of such events. This is particularly true for the large scale flank collapse of volcanic islands, a rare but potentially catastrophic tsunami generator (Ward, 2002)
Perhaps the most famous and extreme example of a landslide generated tsunami is that which occurred in Lituya Bay in southeastern Alaska on July 9, 1958. In this event, an on-land earthquake along the Fairweather Fault induced a subaerial landslide that displaced some 30 million cubic meters of material which collapsed into the head of Lituya Bay displacing a similar amount of water. The subsequent wave reached a height of 524 m above sea level proximal to the landslide with wave heights tapering to ~10 m at the mouth of the bay (Miller, 1960). A submarine landslide is believed to be the primary cause of the 1998 Papua New Guinea tsunami which killed more than 2100 people (Synolakis et al., 2002)
Volcanic and other source mechanisms
Tsunamis can occur as the result of a volcanic eruption if the eruptions causes a caldera collapse and results in the displacement of a large volume of water. Additionally, volcanic eruptions and associated earthquakes can produce landslides or other sub-aerial mass movements which then fall into a body of water causing a wave. The 1883 eruption of Krakatoa in the Sunda Strait between the Islands of Sumatra and Java is the most significant contemporary example of this phenomenon (Winchester, 2005).
Other causes of tsunami or tsunami-like waves include meteorological forcing i.e. ‘meteotsunami’, most commonly through the translation of an area of low atmospheric pressure across a body of water. Since low atmospheric pressure causes a displacement of the water surface, if the translation speed of the pressure anomaly matches the local phase speed of the induced water wave, resonant coupling results in an exponential increase in the water wave height (see Dean and Dalrymple, 1991 p. xx). An example of this effect includes the tsunami-like wave that hit Daytona Beach, Florida on the night of July 3rd 1992 (Sallenger et al., 1995). A final (perhaps literally!) tsunami generation mechanism is that caused by an asteroid impact upon an ocean basin. While the no such event is known to have occurred in human history, attempts have been made to quantify the uncertainty of such an occurrence (Chapman and Morrison, 2000) and the size of the waves it might generate (Ward and Asphaug, 2000).
Tsunami Propagation, Inundation and Runup
Due to the fact that the wavelength (the distance between successive peaks in a series of waves) of most tectonically generated tsunami is usually much greater than the average depth of the ocean basins they propagate as shallow water waves. An important consequence of this is that the speed of propagation is governed only by the depth of the water through the relationship:
c = (gd)1/2 (1)
where c is the wave speed, g the acceleration due to gravity (9.8 m/s2) and d the depth of the water. Taking d as 5000 m (the average depth of an oceanic basin) yields a tsunami propagation speed of 221 m/s (797 km/hr, 495 mi/hr). This simple calculation gives rise to the popular maxim that ‘tsunamis travel at the same speed as a jet airliner’. However, as the water depth decreases, so too does the propagation speed and as a tsunami wave front approaches dry land, it has slowed down considerably relative to its propagation speed in the deep ocean.
It is this slowing of the wave front in shallow water that ultimately leads to the destructive inundation of tsunami waves. Due to their very long wavelengths, the speed of the leading edge of a tsunami wave will approach zero as following sections of the wave are still travelling at full speed. This results in the ‘train wreck’ effect where the advancing wave piles up behind the slower moving wave front. Due to conservation of mass, the water contained in the wave train must increase in height ultimately flooding a coastal area. Depending on the specifics of the tsunami itself and the receiving environment, this flooding can manifest as a rapidly rising water level, similar to a tide (hence the oft-misused moniker ‘tidal wave’), or it can result in violently breaking waves and destructive high-speed inundation. The terms for tsunami inundation and runup are defined in Figure 1.
Figure 1. Definition sketch for tsunami inundation and runup.
Recent Advances in Tsunami Science
Since the early 1990’s there have been significant advances in many aspects of tsunami science and hazard mitigation. This is partially attributable to two tsunamis occurring in 1992. The first, caused by the 25 April Cape Mendocino earthquake at the southern end of the Cascadia Subduction Zone (CSZ) in Northern California, generated only a small, non-damaging tsunami (~1 m) that affected the coast immediately after. This event was important in that it served as a reminder of the potential for locally generated, near-field tsunamis occurring along the US west coast and as a reminder of the seismicity of the CSZ, recently identified as a source of very large earthquakes and subsequent tsunamis (Atwater, 1987). This was followed on 2 September 1992 by a M~7.2 earthquake off the Pacific Coast of Nicaragua. Although the earthquake was relatively small (roughly the same surface wave magnitude of the Cape Mendocino event), it produced a highly localized tsunami with wave heights of up to ~10 m and killed more than 100 people. It was following the Nicaragua tsunami that the modern era of systematic post-tsunami reconnaissance surveys began. This trend has continued since then and has led to a more nuanced and complete understanding of tsunami effects in both the near and far field (Synolakis and Okal, 2005). Around the same time, significant advances in computer hardware made it possible for the first time to begin to accurately simulate the details of tsunami propagation and inundation and the mid-1990’s saw the development of several numerical models designed specifically for this task. In the USA, efforts towards tsunami hazard mitigation also increased in this time frame. In 1997 the US Congress created the National Tsunami Hazard Mitigation Program (NTHMP) which directed the National Oceanic and Atmospheric Administration (NOAA) to develop and lead a program aimed at improving tsunami preparedness in the USA through hazard assessment, warning guidance, and mitigation (Bernard, 1998).
Tsunami Deposits, Paleo-tsunami studies
The study of tsunami sedimentology has advanced considerably in the past 25 years. Since Atwater’s (1987) association between depositional sand layers on the Pacific Coast of Washington State and the occurrence of great earthquakes and associated tsunamis, more and more researchers have integrated sedimentological analyses into post-tsunami field survey efforts. The resulting data base has expanded the understanding of the depositional mechanisms responsible for tsunami deposits and assisted in the identification of tsunami deposits in other parts of the world. Recognizing and identifying tsunami deposits is an important part of peleoseismological studies since deposits can be used to determine the magnitude and extent of prehistoric tsunamis or to verify the occurrence of such an event. In the early 2000’s deposits resulting from an earthquake and tsunami in 869 AD were identified on the Sendai plain in Japan (Minoura et al., 2001) and suggested that this areas was susceptible to large scale tsunami inundation. This unfortunate reality was confirmed 10 years later as the 2011 Tohoku earthquake created a tsunami, which inundated the same area.
Tsunami in Ports and Harbors
The tsunamis of 2004 (Sumatra), 2010 (Chile) and 2011 (Japan) affected ports and maritime infrastructure throughout the Indian and Pacific Oceans. In the 40+ years since the last major trans-oceanic tsunami (the 1960 Chile an 1964 Alaska events), there has been a significant increase in shipping traffic, commerce and related port development which has drastically increased the risk from tsunami exposure to the maritime sector. The tsunami events mentioned above were notable in that the damaging effects of tsunami induced currents and the extended duration of tsunami activity were widely reported and quantified in terms of their effect on ports (Lynett et al., 2012; Borrero and Greer, 2012). These events also illustrated how tsunami induced currents can be damaging and economically disruptive without causing any overland flooding or inundation. This fact suggests that previous hazard mitigation efforts which focused primarily on inundation extent are not sufficient for application to ports and harbors and the activities conducted therein.
Tsunami Warning Systems
Tsunami warning systems (TWS) are designed to detect tsunamis and issue appropriate warnings. The current TWS is administered through a system of inter-governmental agreements and treaties defining areas of responsibility and operational protocols. In the Pacific Ocean, the Pacific Tsunami Warning Center (PTWC), located in Hawaii, monitors global seismic activity and provides warning for the nations of the Pacific Rim. After the 2004 Indian Ocean tsunami, additional warning systems have been established for the Indian Ocean. Tsunami warning systems are under development in the North Atlantic/Mediterranean and Caribbean Seas.
Existing TWS are comprised of two components; 1) networks of sensors for detecting earthquakes and the tsunami itself and 2) communications infrastructure for the dissemination of the warning messages. Earthquakes detected on seismometers located around the world are used to initiate tsunami watches and warnings while ocean based sensors such as the DART® (Deep ocean Assessment and Recording of Tsunami) buoys can directly verify the existence and size of a tsunami (Bernard and Meinig, 2011). The DART tsunameters were first deployed in 1995 and only became operational since the early 2000’s. Prior to this, tsunami warnings were based only on the characteristics of earthquakes and monitoring coastal tide gauge networks. Warning messages are disseminated to the public through a variety of communications channels including radio, fax, telex, SMS and e-mail. Dedicated communication links have been established for first-responders and emergency managers.
While significant advances have been made in tsunami science and hazard mitigation over recent decades, the work is never done. Tsunamis are not as rare an occurrence globally as we may think; in the past 20 years, there have been at least 29 tsunami events responsible for more than 300,000 deaths world wide (Table 1). However, on regional or local scales, tsunamis tend to occur only once a generation or less. Furthermore, as Satake and Atwater (2007) recognized, the ‘great’ tsunamis in human history do not conform to expected geological rules of thumb. This unfortunate reality was illustrated again in 2011 with the great Tohoku earthquake and tsunami, which like the 2004 Indian Ocean event, was not expected along that plate boundary based on historical precedent. If we accept for a moment that hard science might not provide the answers we need in terms of predicting future events, perhaps for the tsunami problem we should not underestimate the importance of enhancing and expanding efforts in public education to maintain awareness of tsunami hazards over generational timescales.
Atwater, B. F., 1987. Evidence for Great Holocene Earthquakes Along the Outer Coast of Washington State. Science, 236: 942 944.
Borrero, J.C. and Greer, S.D., 2012. Comparison of the 2010 Chile and 2010 Japan tsunamis in the Far-field. Pure and Applied Geophysics, DOI 10.1007/s00024-012-0559-4.
Bardet, J.-P., Synolakis, C.E., Davies, H.L., Imamura, F., and Okal, E. A. eds. 2003. Landslide Tsunamis: Recent Findings and Research Directions, Pure and Applied Geophysics: Topical Volume, ISBN: 978-3-7643-6033-7
Bernard, E.N. (1998): Program aims to reduce impact of tsunamis on Pacific states. Eos Trans. AGU, 79(22), doi: 10.1029/98EO00191, 258, 262–263.
Bernard, E., and Meinig, C., 2011. History and future of deep-ocean tsunami measurements. In Proceedings of Oceans’ 11 MTS/IEEE, 19–22 September 2011, No. 6106894, 7 pp.
Chapman, C.R. and Morrison, D., 1994. Impacts on the Earth by asteroids and comets: assessing the hazard. Nature, 367: 33-40.
Dean R.G. and Dalrymple, R.A., 1991. Water Wave Mechanics for Scientists and Engineers, Singapore, World Scientific.
Miller, D.J., 1960. Giant Waves in Lituya Bay Alaska. U.S. Geological Survey Professional Paper 354-c.
Minoura, K., Imamura, F., Sugawara, D., Kono, Y. and Iwashita, T., 2001. The 869 Jogan tsunami deposit and recurrence interval of large-scale tsunami on the Pacific coast of northeast Japan. Journal of Natural Disaster Science, 23(2): 83-88.
Kanamori, H., 1972. Mechanism of Tsunami Earthquakes. Physics of the Earth and Planetary Interiors, 6: 346-359.
Lynett, P., Borrero, J., Weiss, R., Son, S., Greer, D., Renteria, W., 2012. Observations and Modeling of Tsunami-Induced Currents in Ports and Harbors. Earth and Planetary Science Letters, 327-328: 68-74.
Satake, K. and Atwater, B., 2007. Long-Term Perspectives on Giant Earthquakes and Tsunamis at Subduction Zones, Annual Reviews of Earth and Planetary Science, 35: 349-374.
Sallenger, Jr, A. H., List, J. H., Gelfenbaum, G., Stumpf, R. P., Hansen, M., 1995. Large wave at Daytona Beach, Florida, explained as a squall-line surge. Journal of Coastal Research, 11: 1383-1388.
Synolakis, C.E., Bardet, J.-P., Borrero, J.C., Davies, H.L., Okal, E.A., Silver, E., Sweet, S. and Tappin, D.R., 2002. The slump origin of the 1998 Papua New Guinea tsunami, Proceedings of the Royal Society (London), Ser. A, 458: 763-789.
Synolakis, C.E., and Okal, E.A., 2005. 1992-2002: Perspective on a decade of post-tsunami surveys, in: Satake, K. (ed.) Tsunamis: Case studies and recent developments, Advances in Natural and Technological Hazards Research, 23: 1-30.
Ward, S.N., 2002. Slip-sliding away. Nature, 415: 973-974.
Ward, S.N. and Asphaug, E., 2000. Asteroid Impact Tsunami: A Probabilistic Hazard Assessment. Icarus, 145: 64-78.
Winchester, S., 2005. Krakatoa: The Day the World Exploded: August 27, 1883. New York, Harper Collins.