Parkfield quake prediction certified.

Parkfield quake prediction certified

For the first time, an earthquake forecast has been given an official stamp of approval. Dallas Peck, director of the U.S. Geological Survey (USGS), announced last week that a prediction of a magnitude 5.5 or 6 earthquake occurring near Parkfield, Calif., within the next eight years has been endorsed by the National Earthquake Prediction Evalution Council (NEPEC), a scientific review panel established by Congress in 1977 to advise USGS. NEPEC's counterpart in California also gave its blessing to the prediction.

The only other prediction ever presented to NEPEC for evaluation was a forecast for Peru, and this was turned down by the council in 1981 (SN:7/4/81, p. 5). Unlike the Peru prediction, the Parkfield forecast specifies an exact epicenter, magnitude and time window and in this sense represents a fundamental advance in the scientific understanding of the earthquake process--at least in some regions. It is also helping scientists to place equipment around the fault so that they can zoom in on the details of earthquake dynamics and precursors of future quakes.

The Parkfield prediction was originally made last spring by Thomas McEvilly at the University of California at Berkeley, with William Bakun and Allan Lindh at USGS in Menlo Park, Calif. They noted that quakes have occurred on the San Andreas fault near Parkfield about every 22 years, and since the last one struck in 1966, another was due soon. But what really strengthened their forecast was the observation that seismograms painted an essentially identical picture of the last three quakes, indicating that the same part of the fault had ruptured in the same way each time. Bakun adds that historical data on two earlier quakes in 1881 and 1901 are consistent with these seismograms.

All this enabled the researchers to construct a model of a characteristic Parkfield quake and in so doing to make a detailed and precise prediction about a future quake with a 95 percent probability. A previous forecast for Parkfield, based on a statistical analysis of the relative probabilities of quakes along the fault, gave a 75 percent chance that a moderate quake would happen within two decades (SN:12/24/83, p. 404).

Along with its endorsement of the prediction, NEPEC gave qualified support to a suggestion of Kerry Sieh at Caltech in Pasadena and Richard Janhs (now deceased) that a future Parkfield quake might either trigger a second quake to the southeast or itself extend farther in that direction.