TOPICAL COMMENT Seth Stein Æ Emile A. Okal The 2004 Sumatra earthquake and Indian Ocean tsunami: What happened and why? Received: 9 March 2005 / Accepted: 22 March 2005 / Published online: 11 June 2005 Ó Springer-Verlag 2005 Abstract The December 26, 2004 Sumatra earthquake and the tsunami that followed killed over 300,000 peo- ple. In this paper, we analyze and discuss the geologic causes for this earthquake, the mechanisms that gener- ated it, and follow up with a discussion on ways to prevent this type of disaster in the future. Keywords Earthquake Æ Sumatra Æ Tsunami Introduction On December 26, 2004 the world saw, yet again, how strains, built up over hundreds of years by slow and almost imperceptible motions of tectonic plates, are re- leased with devastating effects. A giant earthquake be- neath the Indonesian island of Sumatra generated a massive sea wave that crossed the Indian Ocean in a few hours, wreaking destruction along seacoasts and causing at least 300,000 deaths. Geological cause The geologic causes of this event can be traced back over 120 Myr, when the southern supercontinent of Gondwanaland broke up. The subcontinent of India separated from Antarctica and started its steady motion northward. Fifty million years ago it collided with Asia, raising the Himalayas and forming the Tibetan plateau. The plate collision continues today as the Indian plate moves northward, forcing pieces of China and Southeast Asia eastward. Part of the plate boundary extends along the trench on the west coast of Sumatra. Here, an oceanic part of the Indian plate is being subducted beneath the Burma plate, a small sliver or microplate, between the Indian plate and the Sunda plate that contains much of Southeast Asia (Fig. 1). The east-dipping Indian plate can be identified by earthquakes that occur within it, down to a depth of about 300 km. However, most of the time, little seems to be happening along the megathrust fault that forms the plate-boundary interface. In reality, a lot is happening. Every year, about 20 mm of convergence occurs between the Indian and Burma plates. However, the megathrust fault is locked, so strain builds up on it (Fig. 2). Eventually the accu- mulated strain exceeds the frictional strength of the fault, and it slips in a great earthquake, such as that which occurredin December. Such plate-boundary, thrust-fault earthquakes can be very large and enormously destructive. A huge area of the plate interface slips, generating seismic waves that can cause great damage near the earthquake. Moreover, because this typically occurs at ana marine trench, the overriding plate that had been dragged down since the last earthquake rebounds, and displaces a great volume of water, causing a tsunami that can have devastating effects far awaymany kilometers away. Measuring earthquake size The magnitude of the earthquake has consequences for the slip process and the generation of the tsunami. Different ways of measuring earthquake magnitude use seismic waves of different periods and so provide dif- ferent information, because an earthquake radiates different amounts of seismic energy at different periods. A commonly used representation (Fig. 3) is to plot the spectrum of the earthquake source as the logarithm of the amplitude of the radiated waves versus the loga- rithm of the wave frequency (1 / period). Ideally the plot is flat at low frequency (long period) and then decays for frequencies above (periods shorter than) ‘‘corner’’ frequencies proportional to 1 over the times S. Stein (&) Æ E. A. Okal Department of Geological Sciences, Northwestern University, Evanston, Illinois, 60208 USA E-mail: seth@earth.northwestern.edu Vis Geosci (2006) 10: 21–26 DOI 10.1007/s10069-005-0025-x