People have properly expressed concern over the use, or the misuse, of our growing ability to provide earthquake WARNINGS. I suspect that this concern arises from significant misunderstanding of the basic issues involved:
- Earthquakes rarely kill people. Buildings, and other manmade structures, often kill people when subjected to earthquakes.
- A properly designed warning SYSTEM can sometimes provide DAYS of warning, with ALERTs extending for a week or more. Some seismic precursors can be detected more than a month in advance.
- When people are given adequate information, they tend to act responsibly. It is fear of the unknown that generates panic.
The Chinese success at Haicheng on February 4, 1975 is the classic case illustrating those three principles. Few people were killed or injured because the buildings were evacuated several hours before the earthquake struck. This was possible because a variety of signals were monitored. There was no need to evacuate the area; the few people who refused to leave the buildings paid the price for their stubbornness. (Tributsch; Hazen)
Indeed, ill-advised attempts to evacuate a threatened area could lead to catastrophe, as evidenced by the freeway damage from the Loma Prieta (1989), Northridge (1994), and Kobe, Japan (1995) experiences. If the World Series baseball game had been cancelled by an earthquake warning, the death toll on the I-880/Cypress structure could have been three to five times higher. At Northridge, with the exception of one apartment complex, most people were in a very safe place - home in bed. This was not so true in Kobe, because of differences in building construction.
I suspect that four hours of warning at Kobe, for evacuation of buildings and city streets, could have cut the toll by 50%. A few months later at Sakhalin Island (1995), ten minutes of effective warning may have saved 90% of the victims. The differences between these two cases were the building construction and building/population densities.
These opinions have been gleaned from my studies of the problem, which began in 1982, and from thirty-five years as a firefighter, where I specialized in communications, databases, and water supplies. Additional insight has been acquired from materials provided through FEMA, USGS, Earthquake Engineering Research Institute, and other sources.