Twenty years of multinational earthquake prediction research in Iceland, and the future
As in all earthquake-prone countries it has for a long time been a challenge for Icelandic seismologists to be able to warn about large earthquakes. During 1988-2005 Icelanders coordinated four international prediction research projects, with Iceland as a test-area, to build a base for earthquake warnings and better hazard assessments.
The SIL project (1988-1995) was a concerted effort of the Nordic countries, Denmark, Finland, Iceland, Norway and Sweden in earthquake prediction research. The test area was the earthquake-prone region of the South Iceland Lowland, known as the South Iceland Seismic Zone (SISZ). The path selected for the project was to study the physics of crustal processes leading to large earthquakes. It was considered that small earthquakes, down to magnitude zero, were most significant for this purpose, because of the detailed information, which they provide both in time and space of crustal conditions and ongoing processes. The most significant outcome of the SIL project was the SIL system, which besides evaluating source function and mechanism information carried from below by micro-earthquakes, provided near real-time information, that is used as the basis for an alert system. The alert system has been in operation for almost two decades and is in further development. The SIL system also provides a basis for all the later prediction research projects.
The PRENLAB and PRENLAB-2 projects of several European countries, 1996-2000 (EC supported projects) were a direct continuation of the SIL project, but with a more multidisciplinary approach. PRENLAB stands for "Earthquake prediction research in a natural laboratory". The basic objective was to advance our understanding in general on when, where and how a dangerous earthquake might strike. Methods were developed to study crustal processes and conditions using micro-earthquake information, continuous GPS, InSAR, modeling, detailed fault mapping, earthquake history, and paleo-seismology.
Two large SISZ earthquakes (Ms 6.6) occurred in June 2000. Both were warned for as concerns the place. The second of the two earthquakes had a short-term warning.
A new European project, the PREPARED project started in 2003 aiming to apply the practical experience of the two large earthquakes of the year 2000 in the SISZ, as said "in the context of earthquake prediction research to develop technology for improving preparedness and mitigating risk".
Icelandic seismologists were involved in two other prediction research projects which also had Iceland as a research basis, the SMSITES project (EC project 2000-2003) led by UK scientists and the RETINA project (EC project 2002-2005) led by French scientists. They have also been involved in three other EC projects which have created a significant addition to the prediction projects, the FORESIGHT project (2004-2006), the TRANSFER project (2006-2009), and the SAFER project (2006-2009).
A few among many significant results are: There is pre-earthquake activity (seismicity) ahead of all earthquakes in Iceland larger than magnitude 5. Micro-earthquakes bring the most significant pre-earthquake information from the seismogenic zone. By this information a heterogeneous stress field around an active fault can be mapped in time and space. A new model, explaining the pre-earthquake activity in terms of fluids and crustal strain, has been developed. The model postulates stress modifications by up-flow of fluids from below.
By use of historical evidence, sensitive geophysical observations, evaluations of interseismic and likely pre-seismic processes, and real-time modeling it is probable that all large earthquakes can be predicted in some way. A necessary condition is:
- A real-time monitoring and evaluation system, like the SIL system.
- The merging of all significant information in an early information and warning system (EWIS) for automatic and manual decision making at each stage of the pre-earthquake process, which may last for decades.
- Continuously ongoing modeling of crustal heterogeneities in time and space.
- Organized earth watching by scientists and the administration, with the goal of being able to predict earthquakes. Currently, the most significant part of earthquake prediction work is the development of a well organized earth-watching system, taking into account all the significant results of the 20 years of earthquake prediction research in Iceland and other countries. This has started to some extent, but the work must be accelerated during the next years.