The Why Files The Why Files --

Understanding Quakes

Indian Ocean disaster
Banda Aceh, Sumatra island, Indonesia devastationOn Dec. 26, 2004, the largest earthquake in 40 years rocked the floor of the Indian Ocean, triggering tsunamis that drowned an estimated 285,000 people. Most of the victims lived on the Indonesian island of Sumatra, others lived in India, Sri Lanka and Thailand.

Banda Aceh, Sumatra island, Indonesia, Jan. 5, 2005: A 22-year-old woman sits quietly besides the debris-littered roadside after visiting the devastated house of her grandmother, who was missing since the tsunami. AP Photo/Peter Dejong

To those who try to anticipate earthquakes, the Sumatra-Andaman Islands earthquake served a slice of humble pie. The Sunda Fault, which divides the Burma and Indian plates, was known to be accumulating strain, says Thorne Lay, of the earth sciences department at the University of California, Santa Cruz. Seismologists were expecting a strong rupture in central and southeast Sumatra, but not in the section that caused the December quake.

Lay, who has studied the December quake (see "The Great Sumatra..." in the bibliography), says there had not been any record of great earthquakes or huge tsunamis for northwest Sumatra or the nearby Nicobar and Andaman Islands -- where the giant December quake jolted a record 1,300-plus kilometers of the fault.

The December earthquake was also deadly: uplift of the seafloor, which measured a good five meters, sparked those killer tsunamis.

Andaman sea seen off the coast of India.
The Sunda (or Andaman) Fault divides the Burma Plate from the Indian Ocean Plate. Image: USGS

How big is big?
The Sumatra earthquake is now estimated at magnitude 9.0 to 9.3, but because magnitude is measured on a logarithmic scale, it was several times weaker than the monster that shook Chile in 1960, with an estimated magnitude of 9.5. Although both quakes sparked massive tsunamis, the human toll of Sumatra was vastly larger, since the 2004 tsunamis struck heavily populated coastlines.

But the Sumatra earthquake did set one record: it broke at least 1,300 kilometers of the fault, while Chile busted "only" 1,000 kilometers.

Chile may have neared the maximum power of an earthquake, says Lay. "There probably is an upper boundary, in the sense that the size of an earthquake depends partly on the length of fault that ruptures. One can imagine all of the plate boundary along South American rupturing in a single, magnitude 9.8 to 10 earthquake, but it is extremely unlikely that the whole length would be ready to go at once." Eventually, the rupture will reach a section of fault where a recent earthquake has reduced the stress. Such a section is unable to slide.

Photo shows destruction of town in Chile after an earthquake
Many houses in Maullín, Chile, withstood the magnitude 9.5 Chile earthquake of May 22, 1960. A tsunami sparked by the quake caused most of the damage seen in this 1960 photo. Photo: USGS

Finding fault
Map shows tight distribution of earthquakes in Andaman SeaJust three months after the disastrous Sumatra earthquake-tsunami, a second quake rocked the Sunda Fault, starting just southeast of the December rupture. Although the March slippage did not spark such deadly tsunamis (about 4,000 were killed by local waves), it was still the second-largest earthquake since 1964, and its location next to the December slippage signifies a connection between the two earthquakes, Lay adds.

Blue circles show aftershocks within one month of the Dec. 26, 2004 earthquake at the Sunda Fault. Red spot shows a "swarm" of quakes in a six-day period. Notice the line of volcanoes just inland of the earthquake zone. Green lines show seismic faults. Graphic: USGS

The one-two punch also points up our limited grasp of Earth's guts. Why, for example, didn't the entire fault bust loose in December, producing a staggering magnitude 9.4 quake? While the December quake probably increased the stress on the region that broke in March, Lay says, "we don't exactly know the conditions on the fault that kept the southern region from failing during the December shock," while allowing it to break just three months later.

At any rate, the March quake ruptured the fault down to the edge of a magnitude 9 earthquake that struck Sumatra in 1833. Strain along that section, Lay adds, "had been building for 172 years, but it wasn't quite ready to fail, or the process of slipping of the March rupture did not push it over the edge."

Timeline of earthquakes across northern Turkey.
A series of large earthquakes has been marching across Turkey for almost a century. Could the next one strike Istanbul, population 9.4 million? Map: USGS

Fact is, one quake can lead to another, as tectonic plates play their slow-motion game of bumper cars. And this kind of one-after-the-other slippage helps seismologists assess earthquake hazards. "The best example of this kind of progressive unzipping of a fault zone, of one event after another, is in Turkey, on the Anatolian Fault," says Lay. In the last century, this fault "has ruptured from the eastern side, at magnitude 7.5 to 8.0, and these have marched westward across Turkey."

Is something pulling the trigger on earthquakes?


Megan Anderson, project assistant; Terry Devitt, editor; S.V. Medaris, designer/illustrator; David Tenenbaum, feature writer; Amy Toburen, content development executive

©2017, University of Wisconsin, Board of Regents.