That 18 April was 100 years since the Magnitude 7.9 earthquake that raised San Francisco to the ground and killed more than 3000 is no cause for celebration. Yet it focussed seismologists to commemorate the event, as if that was necessary following hard on the heels of two of the most shattering natural events of the last century. In fact San Francisco created the science of seismology, rocking as it did the most vibrant city in the world’s emerging superpower. It brought the San Andreas Fault into common parlance, and research on that huge and structurally odd fracture – one of the largest transcurrent systems on the continent – played a major role in the development of plate tectonics. In the US, a century of attention to seismic hazards has made it, along with Japan, the leader in attempts to forecast earthquakes and subdivide half a continent in terms of seismic risk (see Here is the earthquake forecast in the July 2005 issue of EPN).

The 1906 San Francisco earthquake is reviewed in issues of three generalist journals (Lubick, N. 2006. Breaking new ground. Nature, v. 440, p. 864-865. Holden, C. 2006. Reliving the ‘Frisco quake. Science, v. 312, p. 345. Marshall, J. 100 years on, you’d think San Francisco would be ready. New Scientist, v. 190 15 April 2006, p. 8-11). In each, different graphics show the estimated risk of earthquakes and the degree of seismic hazard in relation to the many large faults in California. Yet the Sumatra-Andaman earthquake that set the Indian Ocean tsunamis in motion on 26 December 2004, and that in Kashmir in October 2005, between 20 and 40 times more energetic than San Francisco, killed hundreds of times more people and devastated the lives of millions more. As well as more widely deploying well-known, sensible and moderate-cost measures to build and site habitations more safely as regards the shaking effects of seismic waves, a great deal is left to learn about the global nature of earthquake hazard. A first step is better understanding the actual processes to which great earthquakes are related, and lessons are beginning to stem from the research on the Sunda subduction zone, whose movement unleashed terror around the entire Indian Ocean (Briggs, R.W. and 13 others 2006. Deformation and slip along the Sunda megathrust in the great 2005 Nias-Simeulue earthquake. Science, v. 311, p. 1897-1901). The Nias earthquake involved failure of the Sunda subduction zone in a 400 km gap between that affected by the Sumtra-Andaman event of 2004 and a stretch further to the SE that had three great earthquakes between 1797 to 2000; i.e. a previously quieter sector had succumbed to tectonic forces. That emerged from seismic analysis at the University of Ulster (see Yet more Indian Ocean earthquakes? Sadly, yes in the April 2005 issue of EPN). Briggs et al. examined hundreds of patches of coral reef around the islands of Nias and Simeulue, using preciseGPS measurements of the elevation of coral heads that had been uplifted and killed by exposure to the air. Their results show that uplift was as high as 3 metres with some areas subsiding by around a metre, but the total movement by thrusting beneath the islands was of the order of 11 metres.