German Contribution Towards a Tsunami Warning System

R. Kind, GFZ Potsdam, Germany

kind@gfz-potsdam.de
www.gfz-potsdam.de/geofon/

German scientists have a long experience in research fields required for the development of a tsunami warning system. These fields are oceanography (e.g. at IFM-GEOMAR, AWI, GKSS, BGR), space geodesy (e.g. at GFZ, DLR), seismology (e.g. at GFZ, BGR) and geodynamic modelling (e.g. at GFZ), which are the most important scientific components of a tsunami warning system. Also administrative efforts for mitigating environmental hazards have a long history in Germany. I am concentrating here on the seismological contribution towards a tsunami warning system.

The GeoForschungsZentrum (GFZ) Potsdam operates a modern global seismic network (the GEOFON network) of about 45 stations with mostly near real-time data transmission and open access to its archive of many years of continuous data. Most of the software and some of the hardware (SeisComP, SEEDLink) required for recording, transmitting, quality checking, storing, retrieving and processing the seismic data has been developed at the GFZ. The GEOFON system serves now as an example for a number of other networks inside and outside Europe. Data from these partner network are acquired at GFZ together with those from the actual GEOFON network, forming the so-called GEVN (GEOFON Extended Virtual Network). Most GEVN stations are concentrated in Europe. GEOFON is a member of the Federation of Digital Seismic Networks (FDSN) and cooperates closely and exchanges data with other open national or international networks. The goals of the GEOFON project are fast information of the public in the case of damaging earthquakes, and studies of the Earth's deep interior and of the earthquake source. We have developed an automatic global near real-time earthquake

detection and location system. Our earthquake locations are published near real-time in the Internet, and in cases of possibly damaging earthquakes we also send messages as emails and SMS to interested agencies, news media, scientists and private users.

On December 26, 2005 the automatic GEOFON system has located the Sumatra event 12 minutes after the origin time and published the location at the GFZ Internet home page and distributed messages to seismological data centers and other users.

Fig.1: Seismic stations in the region of the Indian ocean.

Fig.1 shows the locations of the available open seismic stations in the region of the Indian ocean. Fig.2 shows the location of GEOFON and other GEVN stations. No stations in the region of the Indian Ocean were included at that time. Also shown are in Fig.2 the isochrones of the travel times of the P wave. The event was located 12 minutes after origin time. The stations within the 12 minutes isochron have been sufficient for the initial location. GEOFON has presumably been the first data center which informed the public about the Sumatra event.

Fig.2: Stations used by GFZ for near real-time location of Sumatra event. Isochrons of P phase are marked.

Fig.3: Estimated travel time of tsunami. Event location was obtained after 12 minutes.

Fig.3 shows the estimated isochrones of the tsunami with the 12 minutes time marked. However the GEOFON information would not have been sufficient for a fast tsunami warning because the body wave magnitude determined by GEOFON did significantly underestimate the true magnitude. All other data centers had similar problems. The true Magnitude of 9.3 was determined only weeks later.

In conclusion it can be said that the existing modern open global seismological network is a good base to describe very large seismic events properly in near real time. However some more work is needed to achieve in full this goal. First, more open seismic stations are required in the Indian Ocean area (or other areas of potential catastrophic earthquakes) in order to collect a sufficient number of seismic waveform data in a shorter time. With such stations, the time to locate earthquakes could be reduced to about 5 minutes in the Indian ocean. Such a densified network would also permit improved research in understanding the geodynamics of the Indian ocean region, which causes these catastrophic earthquakes. Secondly, more seismic research is necessary for reliable automatic determination of earthquake parameters (especially moment magnitude, source extension and slip distribution). These parameters are necessary for reliable prediction of tsunami generation and propagation. German politicians and and scientists are offering cooperation in all fields of a tsunami warning system. In seismology this offer could include cooperation in installing and operating a seismic real-time network of about 40 stations in the Indian Ocean area, including the necessary data centers.