Tsunami Alert System Starts Up in Indonesia

Four years after the mother of all monster waves struck, Indian Ocean countries will know of a possible tsunami just 3 minutes after an earthquake

Indonesia has switched on a tsunami detection system designed to prevent a recurrence of the disaster following the monstrous wave of 2004, which killed at least 130 000 of its people and nearly half that many in other countries.

Waves that big strike the shores of the Indian Ocean only about once every two centuries, on average. But even between such big events, the German-Indonesian Tsunami Early Warning System (GITEWS) should pay for itself by helping to mitigate the effects of relatively small earthquakes and the lesser waves that they produce. The first test came on 17 November 2008, just days after the system went live, when a fairly strong earthquake hit central Indonesia and the government issued a tsunami alert, though no killer wave appeared.

The detection system uses ocean-floor devices that measure changes in water pressure and surface buoys that both relay the information to a satellite network and add their own measurements of changes in sea level. Software then analyzes the data with algorithms calibrated to account for the depth of the water, the shape of the ocean floor, and other factors identified previously during simulations. By thus precalculating aspects of any conceivable tsunami, the system cuts the time it takes to recognize a killer wave. That time is critical, because most quakes in the Indian Ocean occur uncomfortably close to the Indonesian archipelago.

”You have to be very quick, because wherever the earthquake happens, [the tsunami] will hit the coastline in no more than 30 to 40 minutes,” says Jörn Lauterjung, of the German Research Centre for Geosciences, in Potsdam, which designed much of the system. ”We had to develop new seismic analysis programs to reduce warning time from the 10 to 15 minutes that are usual worldwide to just 3 minutes.”

Germany has donated 51 million (US $68 million) to the project, which is between one-third and one-half of its estimated final cost. Indonesia furnished most of the remaining money, about half of which went to establish communications systems and public-education programs. Without such ”soft” technology, extra warning time would be of little value. A case in point was Somalia, where an estimated 289 lives were lost in 2004 because the country was not able to exploit the 8 hours of warning it had of a tsunami’s approach.

Spotting a tsunami is harder than it may seem. The wave starts off in water so deep and at an amplitude so low that it could pass under a small boat without waking its occupants. Only in the shallows near the shore does it stand up as a wall of water on the beach. For that reason, scientists must track several different sorts of information in time and in space.

”You have to find the right waves,” Lauterjung says. ”There’s the primary wave; the secondary waves, which are slower; and then the surface waves, which are slower still. If your instruments are far enough from the earthquake, you can pull the various signals apart, but if you’re nearby, they’ll overlap, making it hard to estimate the magnitude.”

Such fine-tuning isn’t needed in most of the Pacific region, where quake epicenters are so distant that saving 9 or 10 minutes is of no particular advantage. The exception is Japan, which like Indonesia is in an earthquake zone; it came up with its own analysis system years ago.

Why not take the Japanese results and apply them directly? Lauterjung says the conditions are so different that the work had to be done from scratch. Why turn for such advice to Germany, which has nearly no experience with tsunamis?

”Seismology was developed in Germany,” he responds, the pride audible in his voice. ”We also have a long tradition with ocean modeling, tide models, and other such things. So though we’re not dealing with tsunamis, all the different components needed to detect them are already there for us.”