Where Have All the Earthquakes Gone?

Abstract

T he sky turns a seasick shade of green, signaling the impending arrival of a torrential downpour, complete with lightning galore and howling winds. But moments before the first fat raindrops hit the ground, the gale ceases abruptly in a pregnant pause the calm before the storm. Seismologists have their own version of that phenomenon. Called seismic quiescence, the term describes how a particular patch of ground will sometimes stop producing its normal quotient of minor earthquakes just before a big one hits. That's why geophysicists James H. Dieterich and Paul Okubo grew interested when they noticed a lull in tremors on one side of Hawaii's Kilauea volcano during the last few years. The quiescent area sits on Kilauea's eastern flank, just south of the volcanically active east rift zone, according to Dieterich, a researcher at the U.S. Geological Survey in Menlo Park, Calif. Okubo works at the USGS' Hawaiian Volcano Observatory. On its own, a simple observation of quiescence might not capture much attention among seismologists. Because faults behave irregularly from one year to the next, some regions may take a rest for a short spell without ever generating a big quake. Quiescence does not always precede something important. But the Kilauea case stands out because Dieterich and Okubo have discovered that the currently inactive region has a history of going silent before large quakes. The quiescent zone is an irregularly shaped block of crust measuring about 18 kilometers on its longest side and extending to a depth of 10 kilometers. Researchers believe that a major fault cuts through the bottom of that block, forming a nearly horizontal break that separates the upper crust from deeper regions. Riding on that fault, the top layer is slipping toward the ocean with unusual haste. Surveys using Global Positioning System satellites show the southeastern part of the volcano moving seaward at the geologically exceptional rate of 10 centimeters per year. Calculations suggest that at the depth of the fault, the top layer is slipping even faster at 25 centimeters per year, the fastest creep ever detected along a fault (SN: 6/12/93, p.382).