Perspectives on Tsunami Hazard Reduction

A tsunamigenic earthquake occurred on June 3, 1994 with a violent tsunami involving the south-eastern Java coasts (Indonesia). The results of the report of the field campaign undertaken by an international team (Italian, French and Indonesian) during June 22 to June 28, a few weeks after the tsunami occurrence, are described in this paper, integrating the data obtained by a concurrent survey carried out by Asian and USA researchers. During the survey the authors focused their attention on the most affected districts in Java and Bali islands, visiting many coastal villages in order to interview inhabitants, to analyze damage, and to measure tsunami heights and inundation. It was underlined that the earthquake occurred during the night and caused no damage, being felt only by a few people. About 30–40 minutes after the shock generally three waves attacked the coast with documented heights often exceeding 5 meters, causing the destruction of some villages and claiming more than 250 victims. In this paper the authors present the results of their investigations, describing in detail tsunami heights and inundation as well as destructive tsunami effects at coastal villages This paper complements a previous preliminary report on this disaster (Maramai and Tinti, 1996).

As a result of a feasibility study, a concept and a prototype of the Expert Tsunami Database (ETDB) was developed at the Tsunami Research Group of the Novosibirsk Computing Center, Russian Academy of Sciences. The concept of the system is based on the integration of numerical models with observational, historical and geological data, and processing and analysis tools. These components are integrated with the visualization and mapping software embedded in a specifically developed graphic shell providing fast and efficient manipulation of maps, models and data. The ETDB is intended to be a comprehensive source of observational data on historical tsunamis and seismicity in a particular region, along with some basic additional and reference information related to the tsunami problem. The primary collection of data and their long-term storage are made within the Data Base Management System dBASE IV; however, the further retrieval, dissemination, and processing of data are made on the basis of a specially developed graphic shell that is an enhanced environment for IBM PCs and compatibles for data handling and manipulation and can be used independently from the mother’s database. In its current form, the ETDB contains the condensed source data on more than 800 tsunamigenic events occurring in the Pacific between 684 to 1995 and detailed data (extended source parameters, observed heights, original historical descriptions, etc.) for 129 Kuril-Kamchatka events which occurred within the region since 1737 along with basic reference information on regional seismic and mareograph networks, regional geography, geology and tectonics. Additionally, it includes some blocks for tsunami modeling (e.g. calculation of static bottom displacement, tsunami travel time charts) and some standardized built-in tools for data processing and plotting..

On October 4, 1994, at 6:23 am (U.S. Pacific Coast time), a Mw 8.2 earthquake which generated a tsunami occurred off the coast of Hokkaido, Japan,. At 8:05 am, a tsunami warning was issued by the Pacific Tsunami Warning Center. The estimated arrival time of the first tsunami wave to the west coast of the U.S. was forecast to be at 3:30 pm. The warning was reaffirmed throughout the day with varying degrees of certainty. At 3:04 pm (after 7 hours), the warning was canceled.

The National Research Institute for the Earth Science and Disaster Prevention (NIED) started a project for the prediction of earthquakes and tsunamis by an ocean bottom cable network system in the Sagami trench in 1991. The system is composed of 6 seismographs and 3 tsunami gauges and expected to be completed by the end of March of 1996. After the installation of the system, it is expected that the hypocenters of. earthquakes whose magnitude are over 1.5 could be determined very accurately around the Kanto-Tokai area. Tsunami gauges could detect tsunamis with 1 mm resolution in deep ocean due to the thermal noise reduction treatments.

The slip distributions of five great Alaskan-Aleutian earthquakes of the 20th century are determined by inversion of tsunami waveforms recorded on tide gauges. These earthquakes are the 1938 Alaskan, 1946 Aleutian, 1957 Aleutian, 1964 Prince William Sound, and 1965 Rat Islands earthquakes. These earthquakes have in general not been well-studied due to a lack of seismic data. The tsunami waveform inversion gives previously unknown information about the earthquake source. The results of tsunami inversion are compared to seismic inversion results where they exist and show high correlation to the seismic results. The slip distributions show highly variable slip occurred in each earthquake. The areas of highest slip, asperities, may be the locations of future high slip in great earthquakes. This information is vital because asperities control the location of strong ground motion and the run-up height of tsunamis. These results can be used to make predictions concerning future large earthquakes.

A numerical simulations of the 1993 South-West Off Hokkaido Earthquake tsunami was carried out. The simulated tsunami height distribution along the Hokkaido coast supports a fault model of a westward slip with a steep angle rather than that of an eastward slip with a low angle. The mechanism of tsunami flooding onto Aonae district, Okushiri island, was also made clear by the simulation in which shoaling and refraction due to the shallow area south of the island amplified tsunami height and changed its direction from west to southeast. The flooding simulation indicated that a second tsunami attacked Aonae district from the east after the first wave from the west had attacked the south lowland area of the island.

We performed inverse and forward modeling of the 1993 tsunami around Okushiri Island from the Hokkaido Nansei-oki (Southwest off Hokkaido) earthquake, one of the best documented tsunamis by many field survey teams. For tsunami numerical computations, there are three important factors that control the results: initial conditions, governing equations, and output quantities to be compared with observations. In order to accurately estimate the initial water surface displacement, we first conducted inverse modeling from recorded tsunami waveforms on tide gauges. The linear long-wave theory was adopted to simulate the propagation, and the superposition principle was assumed. The tsunami source thus estimated is also supported by seismological and geodetic data. We next performed forward modeling using this water surface displacement as the initial condition, and computed tsunami run-up on Okushiri Island. The non-linear shallow water equations, including bottom friction and moving boundaries, were adopted. The computed results were compared with field observation data, including run-up heights, inundation area, flow velocity and direction. The computation roughly reproduced the inundation area and tsunami flow direction, but underestimated the run-up heights and velocity, showing the limitation of tsunami numerical modeling method.

The historical tsunami that occurred on July 30, 1627 in Gargano (Apulia, Southern Italy) was generated by a very large earthquake (I=XI MCS scale) that produced severe damage in the whole promontory. In spite of the large number of macroseismic observations, it is not possible to determine the epicenter and the generative fault position unambiguously. Though the historical sources concerning the tsunami are not extremely detailed, yet they allow us to locate and to evaluate the most important wave effects on the coasts. A prior tsunami study, conducted by performing numerical simulations based on integrating shallow-water equations via a finite-element technique, assumed a genetic dip-slip focal mechanism on faults that were temptingly placed in different positions both on land and offshore (Tinti and Piatanesi, 1996). The present work, which represents a natural continuation of that investigation, aims mostly at constraining the location of the tsunamigenic fault on physical grounds; this is accomplished by means of tsunami simulations on a new finer finite-element grid and by using the earthquake sources described in the previous paper as well as new inland faults striking N-S. It is shown that mesh refinement leads to better solutions, especially as far as the computation of the maximum elevation along the coast is concerned. Determining the position of the causative fault was one of the main motivations of this research. It has been found that compatibility with the available tsunami data requires that the fault be located inland in the coastal area embracing the Lesina Lake and the mouth of the Fortore River: the strike, however, is not too well constrained, though E-W striking seems preferable to N-S.

Coastal communities require reliable, accurate, and timely warning of approaching tsunamis. Presently, warnings of a potential tsunami rely on the detection of a seismic disturbance with a submarine epicenter. However, the ability to quantitatively predict the tsunamigenic potential from seismic signals has not been demonstrated. Mid-ocean, bottom mounted pressure sensors could provide advance warnings for the coastal areas far from the source (far-field tsunamis) if the signals were available in near real time (F. Gonzales, personal communication). While such sensors represent a vast improvement over current practices, since the generation of a tsunami is verified, the measured mid-ocean amplitude is quite small. The critical information is the eventual amplitude of the tsunami run up at points along the affected coastline; these values can be highly variable spatially. Validated transformation models, using the mid-ocean tsunami signal as input, may provide an effective warning for those situations.

It is found empirically that, at tide gauges in the eastern North Pacific, later waves in a teletsunami are bounded in amplitude by envelopes that can be determined from observations of earlier waves at the same site. This is illustrated using the teletsunami generated by the 4 October 1994 Shikotan earthquake. The envelopes have the form η(t)=Aσ_T in which t is the time since the local onset of the tsunami and σ_T is the standard deviation of the tsunami-band fluctuations that are observed in a time interval T immediately preceding t. For t ≤ 10 hr, T = t+2 hr and T = 12 hr thereafter. Values for the multiplicative factor A = 3.0 and the temporal parameters for T were determined by tuning the envelope to several teletsunamis. In principle, such envelopes can be extrapolated forward in time and added to predictions of background water levels to forecast the maximum elevations of tsunami waves that might occur during the next few hours. However, several issues need to be addressed before such a scheme can be used operationally These include the choice of inundation thresholds that are appropriate for tsunamis and the spatial distribution of gauges needed to adequately monitor tsunami waves in a region.

Data of the Meijii and Showa Great Sanriku tsunamis are analyzed. When a tsunami higher than 5 m hit coastal cliffs, a “thunder-like” sound is generated and heard at distant places: The plunging breaker of a tsunami higher than 4 m generates a sudden “thunderbolt-like” sound which is heard only at the beach of the plunging. When a tsunami higher than 2.5 m with a front in the form of a spilling breaker proceeds in the shallow water, a continuous sound like a locomotive can be heard in the area. Some of these sounds may be used as a precursor of tsunami. An example of the blank spot in the area where sounds were heard is shown.

Since 1941, the Japan Meteorological Agency (JMA) has been involved in tsunami forecasting for Japan, which has frequently suffered from tsunami disasters. However, in the intervening years huge tsunami have occurred again (for example, the Japan Sea earthquake tsunami {1983,5,26 M=7.7}, and the Hokkaido Nanseioki Tsunami {1993, 7,12 M=7.8}),. and have taken many lives The experience with these tsunamis has fuels the Japanese desire to provide a more rapid and accurate tsunami forecast system in order to save lives.

The new Tsunami Warning System is made up of the following three components:

The new seismograph network and the satellite-based dissemination system have been completed. Rapid numerical tsunami model is nearing completion and will be operational in a few years.

In this paper, this new system and procedures are introduced. The paper will focus especially on the rapid numerical tsunami model in detail, because it will be one of the first applications of rapid numerical modeling to tsunami forecasting in the world.

On April 25–26, 1992, a swarm of large earthquakes rattled northern California, injuring 98 people and causing $66 million dollars in damage. The swarm began with a M_s 7.1 thrust earthquake that generated a small tsunami recorded at tide stations in California, Oregon, and Hawaii. This earthquake/tsunami event provided importalt evidence that the Cascadia Subduction Zone is seismically active and that tsunamis are likely from earthquakes occurring along the zone. This new evidence also led the National Oceanic and Atmospheric Administration (NOAA), the U.S. agency responsible for tsunami warnings, to evaluate the state of tsunami preparedness for communities potentially affected.

NOAA conducted workshops on tsunami hazard assessment, tsunami warnings, and public education over a 15-month period. Over 50 tsunami scientists, emergency planners, and educators participated and produced technical reports with findings and recommendations. The most significant findings that emerged from these workshops include: A summary of the three workshops is presented in this report, along with 12 recommendations that participants felt would mitigate U.S. tsunami hazards.

The thirteen preceding papers in this volume were presented at the Seventeenth International Tsunami Symposium held by the International Union of Geodesy and Geophysics (IUGG) Tsunami Commission in July 1995 in Boulder, Colorado. The Symposium was part of the XXI General Assembly of the IUGG and was sponsored in part by the International Association of Seismology and the Physics of the Earth’s Interior (IASPEI) and in part by the Union.