Field of Science

Tsunamis in the geological record

Tsunami deposits are well documented in the Holocene and the Pleistocene, in part by the good accessibility in outcrops to rocks of these epochs or when historic records help to identify areas subjected to tsunamis.
Modern databases list more than 2.000 tsunami events for the lat 4.000 years, most of them recorded in documents and chronologies and others inferred by their geological evidence.
It seems also possible that tsunamis in historic tim
es (after 1700) have found place in myths and oral tradition of the local Indian tribes of the Cascade Range. Based on these stories geologists tried to establish a chronology of events, backed by geological evidence.

Fig.1. Temporal distribution of 2341 tsunami events listed in the database of the National Geophysical Data Center, USA. The database contains the events of the past 4000 years until 2001 AD, from SCHEFFERS & KELLETAT 2003.

However such a database has to be very incomplete, tsunami without greater damage or loss of life are likely to be underrepresented in historic documents, tsunamis wit
h disastrous effects can in contrary became overemphasized and tsunamis occurring in uninhabited regions will not even be noted by humans. With the age of colonization and exploration the known and inhabited zones grow rapidly, and so also the record of large, destructive tsunamis apparently experienced a mayor increase.
However it can be assumed that the actual number, frequency and power of tsunami in such a compilation are still inaccurate and probably underestimated in the past and emphasized in the present the occurrence of strong tsunam
In the geologic record examples of ancient tsunamis are however quite rare. The coastal environment, like flood plains or the estuary of a river, are subject to continues reworking, erosion and deposition, a single event like a tsunami can got destroyed even before it's deposits or traces can became fossilized.
Also in such a complex environment single e
vents tend to became homogenized and amalgamated with the "background" sedimentation, like deposits of the tides or storm events.

In theory a tsunami can produce various geologic evidences in four phases: it can both deposit sediments and erode them during generation, propagation, run up on land and backwash current.
The sedimentologic record of the run up by a tsunami on land is well described by this post at "Trough The Sandglass", especially sand layers, and it´s environmental effects at "paleoseismicity" - however tsunamis can transport and deposits giant boulders (like reef debris thrown on land), these boulders are unlikely to be reworked by normal processes of a coastal environment and have a great potential to become fossilized.
Liquefaction phenomena like sand dikes and intrusion during the earthquake are
preserved in the sediments underlying the soil and tsunami deposits.

Fig.2.Worldwide published distribution of coastal boulders thrown on land as evidence for tsunamis. Historical tsunami and storm wave boulders were defined here as those purporting to show clear depositional evidence based on historical descriptions, direct observations, and analyses of aerial photographs during the historical age (from GOTO et al.2010).

There is also indirect biological evidence to infer the occurrence of a tsunami.

A strong earthquake can cause a displacement of great parts of a coastal area and the land can become inundated by the sea. The salt water soon will kill trees and plants growing on this land. Because dead trees will survive for quite a while as "Ghost forests" the tree stumps can became buried in the sediments of the tidal flat. After the displacement the land can rise upward by the continuing tectonic movements and again became dry.
These changes can be observed in the stratigraphic succession: layers of peat or soil with tree stumps will change suddenly to sand and silt layers deposited by the tsunami and the tides. The plant remains can be dated by the radiocarbon method and are used to produce a chronology of the changes.

Fig.3. Summer in the ghost forest in Alaska and the remains of the town of Portage after the earthquake of 1964 and in the year 1998. In the background of the old photo spruce trees are dying and the high tide covers recently subsided land. In the modern photo still few trunks are standing and shrubs cover the land rebuilt by tidal silt (after BOLT 1995 and ATWATER et al. 2005).
The 1964 Alaska earthquake was a megathrust earthquake that began at 5:36 P.M. on Good Friday, March 27,.1964. Across south-central Alaska, ground fissures, collapsing buildings, and tsunamis resulting from the earthquake caused about 131 deaths.

The remains of the trees provide even a more accurate chronology: The sudden occurrence of the event is proved by the tree rings, a gradual subsidence of the land would produce a different pattern in the rings that the sudden interruption often observed in cedar trees along the North American coast.


ATWATER, B.F.; SATOKO, M.-R.; KENJI, S.; YOSHINOBU, T.; KAZUE, U. YAMAGUCHI, D.K. (2005): The Orphan Tsunami of 1700 Japanese Clues to a Parent Earthquake in North America. U.S.G.S. - University of Washington Press: 144
BOLT, B.A. (1995): Erdbeben - Schlüssel zur Geodynamik. Spektrum Akademischer Verlag, Berlin: 219

DAWSON, A.G. & STEWART, I. (2007): Tsunami deposits in the geological record. Sedimentary Geology 200: 166-183

GOTO, K.; KAWANA, T. & INAMURA, F. (2010): Historical and geological evidence of boulders deposited by tsunamis, southern Ryukyu Islands, Japan. Earth-Science Reviews 102: 77-99

SCHEFFERS, A. & KELLETAT, D. (2003): Sedimentologic and geomorphologic tsunami imprints worldwide-a review. Earth-Science Reviews 63: 83-92


  1. Note that I understand that the 1700 event caused a tsunami in Japan, and while one can not rule out another event elsewhere in the pacific, its likely that the Japanese tsunami was caused by Cascadia. Thus we know the date.
    I call this excellent detective work to find the link.

  2. Yes - On 26. January 1700 contemporary Japanese chronicles describe a surprising tsunami, which caused minor havoc, but was not preceded by an earthquake.

    Based on tree rings and oral tradition in the Cascadia Range research by ATWATER et al. assume that an estimated magnitude 9 earthquake which date to ca 1700 (+- 1-2 years) generated this "orphan tsunami".
    The interesting and complete publication is freely available:


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS