Short-term earthquake prediction: Current status of seismo-electromagnetics

Loss of human lives as a result of earthquakes is caused overwhelmingly by the collapse of buildings within less than a few minutes of main shocks. The most urgent countermeasure consists of two key elements. One is strengthening of weak structures and the other is short-term earthquake prediction. Short-term prediction needs precursors. Although some promising precursors are reported, the prevailing views in Japan and elsewhere are overly pessimistic. The pessimism largely roots in the fact that short-term precursors are generally non-seismic and tools developed for seismology are not designed to detect them. Nonetheless, nationally funded large-scale earthquake prediction programs always emphasize the need to reinforce seismometer networks. They do not take into account the views of those in the science community who point to the importance of non-seismic precursors. While there are well-founded causes to be skeptical, the situation needs to be improved. One reason for skepticism is that the observations of precursors have not yet been perfect enough and another is that some important fundamental aspects of non-seismic precursors are still unresolved. We review some of these problems. (C) 2008 Elsevier B.V. All rights reserved.