Scanning an earthquake
Geologically, Taiwan sits near the intersection of the Eurasian tectonic plate and the Philippine Sea plate. As a result, tiny Taiwan has more frequent earthquakes than most of the rest of the world. As small as it is, Taiwan has nearly 700 seismic monitoring stations, including 109 strong-motion sensors that can transmit data instantly-the highest concentration of such stations on the planet. The number of monitoring stations and the continuous upgrading of our computers and networking equipment are helping Taiwan's RTD process data at ever faster speeds.
"The RTD is in a race with the seismic waves," says Kuo. He explains that while it is nearly impossible for those near the epicenter of a strong quake to flee the danger, earthquake warning systems can deliver warnings to those further away, giving them some number of seconds to prepare before the most damaging seismic waves reach them.
In addition to our many land-based monitoring stations, Taiwan has, like Japan, begun setting up stations on the seafloor.
Kuo says that the construction of a seafloor monitoring system will enable the RTD to more effectively sense earthquakes since about 70% of Taiwan's earthquakes are centered beneath the ocean floor off the east coast. He predicts that when the first of the seafloor stations are in place off the northeast coast in the second half of 2010, they will cut the time it takes to report earthquakes in this area by more than ten seconds while also improving the accuracy of the data.
How does the RTD help when disaster strikes? Shin Tzay-chyn, a deputy director of the CWB with a PhD in seismology, offers rail transport as an example: It currently takes the RTD 35 seconds to report quake data. If another Chichi earthquake were to strike, rail lines would be twisted and trains derailed in nearby Hsinchu, Taichung, Changhua, and Yunlin before a warning could be issued. However, the RTD warning would arrive in more distant Taoyuan and Tainan in time to enable emergency evacuations and to bring trains to a halt, effectively keeping them out of the most severely hit areas and avoiding derailments.
From report to warning
The RTD generates earthquake data nearly instantly. But this data is useless unless it can be very rapidly disseminated to critical infrastructure, public spaces, and even individual citizens who are in imminent danger. To be effective, the system must deliver its information in time to give people the crucial seconds they need to reach a safer location.
"Our next step is to use the RTD's strengths to build a robust earthquake warning system," says Chen Liang-chuan, director of the National Science and Technology Center for Disaster Reduction (NCDR).
He notes that in the case of the Chichi earthquake, it took the seismic waves 44-45 seconds to travel from Chichi, Nantou County, to Taipei. Chen says that if we had, for example, been able to transmit an earthquake warning to the residents of the Poshih Homes community in Hsinchuang, Taipei County, and of the Tunghsing Building in Taipei City, they may have had time to flee and need not have died when their buildings collapsed.
Japan, which is also earthquake prone, rolled out the world's first early warning system in October of last year. Close cooperation between government and private-sector broadband networks, communications systems, and broadcast media enable the system to deliver first-hand earthquake data to government bodies, nuclear power plants, utilities, and transit authorities, as well as to places where people gather in large numbers-hospitals, TV stations, major corporations, department stores, and shopping centers. It also notifies the public at large via their cellphones.
The system was put to use for the first time in mid-June, when a 7.2-magnitude quake struck northeastern Japan. Sendai, 100 kilometers from the epicenter, received a warning about five seconds before seismic waves reached it. Railway authorities immediately initiated an automatic shutdown of their system. As a result, the 11 trains in transit were slowing when the waves struck, significantly reducing quake casualties. In Tokyo, some 500 km from the epicenter, NHK was able to flash a warning on screen more than ten seconds in advance of the seismic waves.
Chen says that, drawing on Japan's experience, the NCDR will work with the CWB and the private sector to create an earthquake warning system that meets Taiwan's needs. He anticipates that preliminary plans will be forthcoming by the end of this year.
But Chen is quick to note that Taiwan differs from Japan both in its geography and in the attitudes of its public: Japan is 11 times the size of Taiwan, meaning that it can provide greater lead times on the warnings than Taiwan will be able to. The Japanese public is also more willing to participate in disaster preparedness exercises and more tolerant of false alarms.
For example, Japan's current earthquake warning system is only 68% accurate. Many areas have experienced false alarms, or gotten notices only after seismic waves have struck. In addition, the sensors don't always respond correctly; there have been problems with alarms sounding when they shouldn't and not sounding when they should. If similar issues arise in Taiwan, they might well set off a panic and would certainly result in media finger-pointing.
Chen believes that if a warning system is to be effective, it must be tailored to the disaster preparedness, evacuation, and response training of participating organizations and even individual members of the public. "Otherwise," he argues, "the public will panic and have no idea what to do. In that event, giving them more time would make no difference whatsoever."
How should we develop an earthquake warning system that is appropriate to Taiwan? Shin boldly suggests that the way to provide our small island with warnings is to emulate the old air-raid warning system. When seismic waves of intensity 5 or greater are on the way, we could broadcast warnings over urban PA systems telling people to get to safety immediately.
"It's a simple, easily understood approach," says Shin. "If we design it well, it should greatly reduce the number and severity of casualties."
We still have a long way to go before we'll be able to warn of an earthquake before it happens. We can, however, move towards the next best thing-winning a few crucial seconds in which to save lives and prepare a response. Given that we live on an earthquake-prone island, it is clearly our duty to do so.