Two events during 1998, one in the real world and one in Hollywood, have shaped the current popular perception of tsunami and given them a "split image" which fails to describe the true importance of this natural hazard to the insurance industry.

The first event was an earthquake off the northern coast of New Guinea on the night of 17th July. This triggered a series of tsunami, waves up to 15 m high that crashed into coastal villages along a 25 km stretch of coastline and killed around 2000 people. Television and newspaper coverage over the following days created the impression of tsunami as yet another of the natural disasters that yearly afflict the inhabitants of far away third world countries.

The second "event" was the release of the annual summer disaster movies, which in 1998 took as their theme the effects of the impact of giant asteroids on Earth. Taking their cue from the studies of geologists, astronomers and other researchers into the geological, evolutionary and environmental effects of rare but catastrophic events such as the impact at Chixuculub in Mexico that ended the reign of the dinosaurs 65 million years ago, these films presented images of mile - high tsunami as just one of the spectacular components of civilisation - destroying disasters.

Neither of these images of tsunami would suggest an urgent need for improved understanding of the phenomenon in the workings of the insurance industry. Events such as the Papua New Guinea tsunami, however tragic, have little impact upon the insurance industry because the insurance losses caused by disasters in such undeveloped countries are very small. On the other hand, there is little point in attempting to stimulate interest in insurance against asteroid or comet impacts that occur on average only at intervals of hundreds of thousands to millions of years but have the capacity to wipe out most of the human race when they do.

Between these end - member examples, however, there is a wide spectrum of tsunami caused by different geological phenomena ranging from earthquakes to submarine sediment flows, rockfalls, volcanic eruptions, landslides and smaller but more frequent meteorite impacts. Historical and geological records indicate that these phenomena can producing tsunami ranging from the barely perceptible to waves fifty to a hundred metres high and capable of destroying everything up to ten kilometres inland from the coastlines that they impact. Furthermore, tsunami are not "third world" disasters: the word itself is Japanese, and major tsunami capable of causing thousands of deaths strike the Japanese coastline on average every few decades. Other tsunami, produced on the opposite side of the Pacific ocean, have propagated nearly half way around the world and caused near - simultaneous tsunami disasters in places as widely separated as Japan, Hawaii and California. Nor is the Pacific rim the only region to have been affected by tsunami: a tsunami generated by an earthquake in 1755 destroyed large parts of Lisbon and Cadiz, and geological evidence suggests that collapsing volcanoes in the Canary Islands have generated tsunami which crossed the Atlantic and hurled 2000 ton boulders up 30 metre high cliffs in the Bahamas. In this century alone other tsunami have occurred in places ranging from Norwegian fjords to Greek islands. Tsunami have impacted what are now highly - developed coastlines in the past and will do so again in the future: the only difference is that their effects will extend much further inland than their physical extent, because these coastal areas now contain infrastructure elements such as port facilities and power plants whose destruction or damage can cause massive indirect economic losses.

This book seeks to introduce the insurance industry to the complexity of tsunami and the wide range of phenomena that can cause them, and to explore the implications for estimation of tsunami hazards, frequency - magnitude distributions and evaluations of the direct and indirect insurance risks that they present. We look at subjects ranging from how tsunami are generated and propagate across the oceans; to the mechanisms by which they cause damage when they make landfall; to the means by which disaster planning can reduce the economic losses that result; and to the sources of post - disaster information and mapping which can be consulted to validate tsunami - related insurance claims.

The processes by which tsunamis are generated imply that major destructive events are relatively rare. Thus, during any individual decade it is unlikely that more than a very few major events take place worldwide. However, when considered at the century timescale, the frequency of destructive tsunamis is proportionally increased. Certainly recent history has coincided with a number of destructive events. Probably one of the most graphic eye-witness accounts of a tsunami was by Lieutenant Billing on the US Postal steamer Wateree that on August 8th 1868 was transported onshore by a large tsunami generated by an offshore earthquake at Arica, Chile. He described how the tsunami engulfed the ship that was subsequently "…buried under a half-liquid, half-solid mass of sand and water…", that the ship was transported two miles inland and that the wave "…had carried us at an unbelievable speed over the sand dunes…" and that "…the town had disappeared and where it had stood there stretched an even plain of sand…" (Myles, 1985).

In Europe the coastal areas of the Mediterranean have been particularly susceptible to tsunamis generated by major earthquakes. In the following section four of the most destructive tsunamis that have taken place in Europe since the end of the 17th century are described. Each was associated with considerable loss of life. Each area presently coincides with a region that is extremely popular for tourists. Clearly we need to be aware that major tsunami disasters have taken place previously in such areas.