| |
Project 10: Tsunami modelling and prediction
Project Manager: Carl Harbitz (NGI)
  Introduction
Tsunamis (large waves formed by rapid
mass movements) are a secondary effect of geological events like slides
and rock falls in fjords and lakes, submarine slides and earthquakes.
Tsunami events in Norway are most often due to slides. Other triggering
mechanisms are relevant because they generate similar hazards and
require similar solution approaches. Tsunamis are of major interest
internationally.
Tsunamis have been reported since ancient times, and have caused
thousands of deaths and severe destruction worldwide. The Storegga
slides on the Norwegian Continental Slope have been identified as being
responsible for huge tsunamis in Holocene times affecting Norway,
Scotland and the other European Coasts. In the 1990’s, four tsunamis
ravaged the coast of Nicaragua, Indonesia, Japan and Papua New Guinea
causing loss of 4 000 lives. In Norway, the three most severe known
events, leading to the deaths of 174 people altogether, did all occur in
the twentieth century (Loen 1905, 1936; Tafjord 1934).
Hence, tsunamis constitute a serious natural hazard for the environment
and populations in exposed areas. Future catastrophes can be mitigated
or prevented by tsunami hazard evaluation from statistics and geological
analysis, by risk analyses from studies of slide dynamics, tsunami
propagation and coastal impact, and by tsunami warning systems. Moreover,
tsunami predictions are fundamental in engineering design and location
of coastal installations, dams, submerged bridges, offshore
constructions, aquaculture, etc.
Figure 1: The tsunami generated near Sumatra on 26
December 2004, ravaging the coasts in South and South-East Asia, is the
most devastating tsunami in several hundred years.
Scientific objectives
The project has the following main goals:
- Produce general and powerful tools for
tsunami modelling and combine these tools in a general domain
decomposition framework.
- Investigate hydrodynamic models to
establish the limitations of current tsunami models and to test their
sensitivity to the input parameters.
- Undertake comprehensive scientific
investigation of the performance of domain decomposition and parallel
computing in a tsunami context.
- Establish operational hydrodynamic
models for tsunamis and tsunami generation with coupling to the
geological trigger.
- Improve the status of operational
tsunami modelling at NGI and other Norwegian institutions.
- Perform important case studies on
tsunamis.
- Select criteria for slide and tsunami
hazard and risk evaluation in Norway, considering sliding potential
and consequence of sliding for different triggers and types of slides;
develop slide and tsunami hazard and risk mapping procedures.
Ongoing activities
The Mjølnir Impact and Tsunami
The 140 Myr Mjølnir structure is a marine asteroid impact which was
detected in 1993, in the Barents Sea north of Norway. The simulation of
the impact itself and the early phase of generation have been studied by
Valerij Shuvalov at the Russia Academy of Science. However, the
perspective of ICG has been to investigate the propagation of this
tsunami over large distances across the Paleo Barents Sea. A number of
numerical techniques have been used, including Boussinesq equations,
full potential theory, and ray theory for solitary waves. Moreover,
advanced numerical techniques as domain decomposition and
parallelisation have been utilised, enabling the requested numerical
resolution of short crested undular bores over large areas.
Figure 3:
The maximum surface elevation of the tsunami generated by the Mjølnir
impact. The colourbar indicates the maximum surface elevation in km.
Tsunamis generated by rockslides and the Åknes/Tafjord project
This project deals with the investigation and monitoring at the Åknes
failure in western Norway. This includes geological and geophysical
investigations, drilling with instrumentation and logging, installation
of different monitoring systems, laboratory experiments, simulations of
potential future slide generated tsunamis, development of improved
models for tsunamis generated by rockslides and improved models for
tsunami propagation.
Tsunamis generated
by submarine slides
In 2005, two historical submarine slide generated tsunamis in Norway has
been investigated, the 1888 Trondheimsfjorden slide tsunami, and the ca.
10000 BP Boknafjorden slide tsunami. In both cases, emphasis has been
given on reproducing the tsunami generated by submarine slides. For the
case of the Boknafjorden slide tsunami, this event has been presented as
a possible reason for the extinction of a Stone Age settlement in south
west Norway. For 2006 and 2007, further studies of the Holocene Storegga
slide generated tsunami, and the historical Yermak slide north of
Svalbard is planned.
Figure 4: Snapshot of the
wave generated by the Boknafjorden slide, 2.5 min after slide release.
The colourbar indicates the surface elevation in m.
Links to external
studies
An important objective is
the cooperation with other projects and programs. At present, the
following projects are closely linked to the ICG tsunami project.
Tsunami Risk Reduction
Measures
The 26 December 2004 tsunami caused of the order 200,000-300,000
casualties and devastated large areas along the coastlines of Indonesia,
Thailand, Myanmar, Sri Lanka, India, the Maldives and even some parts of
the east African coast. This project was launched in May 2005 as an
8-month fast-track study to assist the authorities in Thailand with
development of plans for how to deal with the future tsunami risk in a
short term as well as in a long term perspective. The project have been
funded by the Royal Norwegian Ministry of Foreign Affairs, for the
Coordinating Committee for Geoscience Programs in East and Southeast
Asia, in response to a request from the Department of Mineral Resources
under the Ministry of Natural Resources and the Environment in Thailand.
Figure 2: The surface
elevation of the simulated 26. December Indian Ocean tsunami just before
it reached the coast of Thailand. The colourbar indicates the surface
elevation in m.
Modelling of the 26. December 2004 Indian Ocean Tsunami
With respect to the general scientific understanding, the public need
for information, and to develop plans to deal with future tsunamis, a
large part of ICG’s
tsunami activity has been the modelling and understanding of
this huge event. Moreover, the
effect of potential future tsunamis generated by earthquakes along the Sunda Arc have been given a large emphasis, to provide a basis for
consequence analysis and risk assessment of tsunamis in the future for
this region.
Instability of Long wave models
Boussinesq equations are frequently encountered when studying tsunamis
as well as coastal wave propagation, including the effects of
non-linearity and dispersion to the effects on wave propagation.
However, several of the Boussinesq models included in the literature,
proves to be intrinsically instable when the effect of variable depth
are included, and the instabilities are not always caused by the
numerical formulations, but by formulation of the equations. Moreover,
such instabilities have rarely been studied. An important focus of ICG
has been to investigate stability properties of a broad range of
Boussinesq models, by developing a general procedure for model stability
investigations.
Euromargins
This project is organised under the European Science Foundation (ESF)
and aims at an improved understanding of continental margin slope
stability in different geological settings along the European margin,
from the river-fed depositional systems along the south-European margins
of the Atlantic and the Mediterranean, to the high Arctic north of
Svalbard. Research on correlation between geological and geotechnical
parameters and on the effect of gas hydrates on geotechnical properties
form integrated parts of ICG activities.
Tsunamis generated by asteroid impacts, rockslides and landslides (BeMatA):
Granted by the Research Council of Norway, the main focus of this
project is the development of improved methods for tsunami modelling.
The project has granted 1.5 PhD positions on tsunami modelling.
Personell
Carl B. Harbitz – project
manager
Geir K. Pedersen
Peter Gauer
Unni Eidsvig
Bjørn Gjevik
Atle Jensen
Oddvar Longva
Hilmar Bungum
Conrad Lindholm
Sylfest Glimsdal
Finn Løvholt
PhD students
Bård Romstad
Key Contacts
Hans Petter Langtangen
|
