Deposits of tsunamis and their recognition
by
Brookfield, Michael
Land Resource Science, Guelph University, Guelph, Ontario N1G 2W1, Canada

Tsunami waves are produced by any process, which vertically displaces the sea surface. Such processes, accompanied by earthquakes, include direct displacement by extraterrestrial impacts, and displacements of the sea floor caused by explosive submarine eruptions, fault movements, and submarine slides. Unfortunately, for practical reasons, studies of modern tsunami deposits are almost entirely confined to coasts, while studies of ancient tsunami deposits (what few there are) are restricted to preserved offshore facies (Dawson, 1999). Concepts based on coastal observations have to be applied with little control to ancient tsunami deposits (Bryant, 2001). Tsunamis of the order of metres in amplitude are relatively common in all oceans, and at least 50 large tsunami deposits (> 5 metre runup height) occur on the Pacific coast of Kamchatka over the last 7,000 years (Pinegina et al., 2003). More than 2,000 tsunami events have been recorded during the past 4,000 years, some with run-up heights of over 100 metres (Scheffers and Kelletat, 2003) (fig. 1). Fig. 1: Reliable runup values of worldwide tsunamis within the last 400 years (from Scheffers and Kelletat, 2003) Coastal and shallow-water tsunami deposits should thus be common in the stratigraphic record (Coleman, 1968), though very few have been described (e.g. Massari and D’Alessandro, 2000; Takashimizu and Masuda, 2000). Tsunamis capable of affecting deep shelf and oceanic sea floors, however, need to be of very large amplitude and wavelength since wave motion extends only to a depth of about one half the wavelength, and lateral motion is dependent on the wave amplitude. Most open ocean tsunami waves have the required wavelengths (up to 100 kilometres of so) but not the amplitude (a few metres maximum) to significantly move sediment in deep water (Yeats et al., 1997). Distinguishing small tsunami deposits from deposits of large waves associated with storms and hurricane sill is difficult, if not impossible. However, neither storms nor hurricanes are capable of generating waves of magnitudes greater than a few tens of metres. The term ‘giant tsunami’ is used here because it is suitably imprecise. We do not know the actual heights of the waves, and the frequently used term ‘mega-tsunami’ is positively misleading. “Mega’ in physical terminology refers to 106 increase, but since the giant tsunamis considered here are defined on wave height, not energy, they should be called deka- (101), hekto- (102) or kilo- (103) tsunamis. Great earthquakes generate only moderate tsunamis, since an instantaneous 10-metre vertical fault displacement cannot initially generate higher tsunami waves, though they often do trigger large slides (Ben-Menakem and Rosenman, 1972). Large submarine eruptions generate only deka-scale tsunamis: for example, the great submarine Krakatoa explosion (1883) triggered tsunamis only locally 30 meters high in shallow water (Winchester, 2003). Large meteorite impacts into oceans can immediately displace the entire water column and generate tsunamis initially higher than the depth of the ocean. A large kilometre-sized asteroid impacting a deep ocean generates initial tsunami waves of comparable height to the ocean depths, and these would still be more than 100 metres high when they reached the surrounding coasts (Hills and Mader, 1997). The Eltanin asteroid impact (around 1 - 4 km diameter) impacted the south Pacific in the late Pliocene (around 2.15 ma) depositing ejecta and reworking deep-marine sediments into distinctive beds which can be used to characterize such deposits (Gersonde et al., 1997). The end Cretaceous Chixulub impact in Yucatan almost (though less) certainly triggered giant tsunamis, which deposited chaotic and graded deposits in shallow water in Texas, and in deeper water in Cuba and Haiti (Bourgeois et al. 1988; Takayama et al., 2000). Submarine landslides in shallow water can also generate tsunamis nearly as high as the initial displacement if the slide is fast enough (Ward, 2001, Todorovska et al., 2002). Giant tsunamis were observed in Lituya Bay, Alaska, in 1958 where a landslide into the bay generated an initial wave over 500 metres high, which then rapidly dissipated to tens of metres down the 12 kilometre long bay where it sank several fishing boats (Pararas-Carayannis, 1999)(fig. 2). Fig. 2: Lituya Bay tsunami In 1929, a magnitude 7.2 earthquake triggered a huge submarine slump (200 km3) on the Grand Banks of Newfoundland, which, in turn, generated a tsunami that crossed the Atlantic and was recorded in western Europe. The tsunami reached amplitudes of 2 to 7 metres along the Newfoundland coast, deposited a 5 centimetre layer of sand over coastal peats, and killed 27 people in the worst tsunami disaster in Canadian history (Bornhold et al., 2003). Fig. 3: Inferred tsunami waves generated by a giant Canary Islands slide. The slump transformed into a turbid flow over time, moving up to 1500 km eastwards at speeds of up to 100 km/hr (Heezen and Drake, 1964). Enormous mass flows occur from oceanic islands and continental margins. Dozens of major submarine landslides have been discovered on the flanks of the Hawaiin ridges and are among the largest on Earth, with lengths greater than 200 kilometres and volumes of 5,000 cubic kilometres (Moore et al., 1994). These huge Quaternary submarine landslides, and similar slides around the Canary Islands must have generated large tsunamis, though so far, little convincing evidence has been found of them (Keating and Helsey, 2002)(fig. 3). However, the Quaternary second Storegga slide (1,700 cubic kilometres; 7,000 yr BP) off Norway has already been implicated in widespread tsunami-deposited subaerial sands in Norway and eastern Scotland (Bondevik et al., 1997; Dawson et al., 1988). Impact tsunamis are relatively rare events though their deposits should be at least sporadically found in the stratigraphic record. In contrast, very large slides from oceanic islands and continental margins are relatively common (Ward, 2001). Slide-generated tsunami deposits should thus be a normal and common feature in the stratigraphic record and can be modeled in some detail if the area, volume and rate of movement of the slide can be given (fig. 4). Fig. 4: Computer modeling of slide-generated tsunamis. Tsunami deposits can be recognized primarily as rapidly deposited, tabular and extensive unusually coarse layers laid down (or at least) reworked by traction currents within finer grained sections: but, as noted above, it is very difficult to separate these from deposits of storms and hurricanes. In coastal areas, very coarse pebbly and sandy deposits within finer-grained lagoonal or marsh deposits are characteristic of both. For example, the July 17, 1998 Papua-New Guinea tsunami deposited gray-colored sand typically overlying a brown, rooted soil more than 650 metres inland (fig. 1). Very rapid deposition was indicated by plants found bent over and buried by the sand, or removed by the tsunami, leaving an erosive base to the deposit: the lower part of the tsunami included rip-up clasts of the underlying muddy soil. Little internal structure was found, although in a few places some faint horizontal stratification was observed at the top. Overall, the tsunami sand was relatively uniform in thickness (5-10 cm), extended from 60 to 675 m inland, and became finer landward. Local variations in the thickness of the sand deposit were associated with small local topographic variations In shallow marine environments, there are the same problems of separating tsunami from storm and hurricane deposits. At Hornitos, northern Chile, a 7-10 m thick conglomerate bed occurs within a succession of shallow marine sandstones is interpreted as a tsunami deposits and has an erosional contact with underlying strata (Hartley et al., 2001). Large basement boulders (5 m) are angular to very angular and are set within a matrix of very poorly sorted fine to very coarse-grained shell-rich sandstone. The angular clasts indicate limited transport and no marine reworking prior to deposition. And represent alluvial fan sediment incorporated into the bed during backflow. Very well rounded granodiorite and shallow marine sandstone pebbles (maximum 10 m) are shallow marine and shoreline material ripped up during the seaward passage of the tsunami across the shoreface. In deeper water environments, curiously, tsunami deposits are probably easier to recognize, since only very large waves can move (or rework) coarser sediments there. However, at present, few deep-sea deposits can be directly related to a generating tsunami. Also, it is difficult to core such coarse deep-water Recent sediments and so recourse has to be made to ancient deep-sea sediments now tectonically emplaced on land. Unfortunately, such sediments have to be recognized as tsunami deposits from basically theoretically inferred characteristics. Some of these critical characteristics are: coarse, shallow-water derived breccias and sands showing mass flow and wave reworking within finer sediments, and showing progressive reduction in energy. In northeastern Mexico the Cretaceous-Tertiary boundary has as much as three meters of coarse clastic layers in stark contrast with the deep water fine marls deposited above and below it (Alvarez et al.,1992). In the oceanic Jurassic nappes of Oman, a particularly widespread unusual deep-water breccia-sandstone unit contains clasts derived from shelf and nearshore environments, shows wave-reworking, and can be related to massive submarine slides on the contemporary shelf margin (fig. 5). Fig. 5: Graded, wave-reworked breccia in deep-sea Jurassic fan deposit of Oman. Alvarez W., Smit J., Lowrie W., Asaro F., Margolis S.V., Claeys P., Kastner M. and Hildebrand A., 1992. Proximal impact deposits at the Cretaceous-Tertiary boundary in the Gulf of Mexico: A restudy of DSDP leg 77 sites 536 and 540, Geology, 20, 697-700. Ben-Menahem A. and Rosenman., M., 1972. Amplitude patterns of tsunami waves from submarine earthquakes. Journal of Geophysical Research, 77, 3097-3128. Bondevik, S., Svendsen J.I., Johnsen G., Mangerud J., and Kaland P.E., 1997. The Storegga tsunami along the Norwegian coast, its age and runup, Boreas, 26, 29-54. Bornhold B.D., Fine I.V., Rabinovich A.B., Thomson R.E. and Kulikov E.A., 2003. The Grand Banks landslide-generated tsunami of November 18, 1929: analysis and numerical modeling, Geophysical Research Abstracts, 5, 1775. Bourgeois J., Hansen T.A., Wiberg P.L. and Kaufmann E.G., 1988. A tsunami deposit at the Cretaceous-Tertiary boundary in Texas, Science, 241, 567-570. Bryant E., 2001. Tsunami: the underrated hazard, 320pp., Cambridge Univ. Press, N.Y., 2001. Coleman P.J., 1968. Tsunamis as geological agents, Journal of the Geological Society of Australia, 15, 267-273. Dawson A.G., 1999. Linking tsunami deposits, submarine slides and offshore earthquakes, Quaternary International, 60, 119-126. Dawson A.G., Long D., and Smith D. E., 1988 The Storegga Slides: evidence from eastern Scotland for a possible tsunami, Marine Geology, 82, 271-276. Gersonde R., Kyte F.T., Bleil U., Diekmann B., Flores J.A., Gohl K., Grahl G., Hagen R., Kuhn G., Sierro F.J., Volker D., Abelmann A. and Bostwick J.A., 1997. Geological record and reconstruction of the late Pliocene impact of the Eltanin asteroid in the Southern Ocean, Nature, 390, 357-363. Hartley A., Howell J., Mather A.E. and Chong G., 2001. A possible Plio-Pleistocene tsunami deposit, Hornitos, northern Chile. Revista geologica Chile, 28, 117-125. Heezen B. and Drake C.L., 1964. Grand Banks slump, American Association of Petroleum Geologists Bulletin, 48, 221-233. Hills J.G. and Mader C.L., 1997. Tsunami produced by the impacts of small asteroids, Annals of the New York Academy of Sciences, 822, 381-394. Keating B.H. and Helsey C.E., 2002. The ancient shorelines of Lanai, Hawaii, revisited, Sedimentary Geology, 150, 3-15. Massari F. and D’Alessandro A., 2000. Tsunami-related scour-and-drape undulations in Middle Pliocene restricted-bay carbonate deposits (Salento, south Italy), Sedimentary Geology, 135, 265-281. Moore J.G., Normark W.R., and Holcomb R.T., 1994. Giant Hawaiian Landslides, Annual Review of Earth and Planetary Sciences, 22, 119-144. Pararas-Carayannis G., 1999. Analysis of mechanisms of tsunami generation in Lituya Bay, Science of Tsunami Hazards, 17, 193-206. Scheffers A. and Kelletat, D., 2003. Sedimentologic and geomorphologic tsunami imprints worldwide - a review. Earth-Science Reviews, 63, 83-92. Takashimizu Y. and F. Masuda, 2000. Depositional facies and sedimentary successions of earthquake-induced tsunami deposits in Upper Pleistocene incised valley fills, central Japan, Sedimentary Geology, 135, 231-239. Takayama H., T. Tada, T. Matsui, M.A. Iturralde-Vinent, T. Oji, E. Tajika, S. Kiyokawa, D. Garcia, H. Okada, T. Hasegawa, and K. Toyoda, K., 2000. Origin of the Penalver Formation in northwestern Cuba and its relation to K/T boundary impact events, Sedimentary Geology, 135, 295-320. Todorovska M.I., A. Hayir, and M.D. Trifunac, M.D., 2002. A note on tsunami amplitudes above submarine slides and slumps, Soil Dynamics and Earthquake Engineering, 22, 129-141. Ward S.N., 2001. Landslide tsunami, Journal of Geophysical Research-Solid Earth, 106, 11201-11215. Winchester S., 2003. Krakatoa: the day the world exploded August 27, 1883, 432 pp. Harper Collins, New York. Yeats R.S., K. Sieh, and R.A. Clarence, R.A., 1997. The Geology of Earthquakes, 568 pp., Oxford University Press, New York.

Date received: November 24, 2003

Copyright © 2003 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # camu-08.