Quake Theories Get Rattled by Scientists : Geology: Challenge to traditional views suggests that rather than massive buildups of stress, more subtle circumstances may trigger temblors.

An increasing number of scientists, bolstered by data available since the Loma Prieta and Whittier Narrows earthquakes, are challenging traditional ideas about faults and how earthquakes occur. The new theories suggest that reliable quake prediction may prove unachievable in the near future.

Scientists at the U.S. Geological Survey in Pasadena, Harvard University and the University of Nevada at Reno are among those whose research has cast a shadow over optimism among some experts that accurate predictions may come soon.

Rather than massive buildups of stress in rocks that is periodically relieved by great earthquakes ripping through the length of the quake zone, they view quakes--triggered by subtle stress changes--proceeding in a kind of wave until an obstacle stops them.

According to this thinking, the difference between a big quake and a small one may depend more on what kind of impediments there are in these fault zones than on detectable surface stress.

The new concept has triggered a debate between its proponents and traditionalists who maintain that most earthquakes can still be explained by well-known and accepted seismological models.

The traditional concept of earthquakes has been that stress builds up along a surface fault until it becomes so great that the sides of the fault begin to slide against one another.

Traditionalists argue that one could determine an average recurrence interval by calculating how long it takes the stress to mount to the critical point. As more precise stress calculations are made and a better understanding of foreshocks and other precursors develop, the predictions would become increasingly accurate, according to the traditional theories.

In keeping with this, scientists in 1985 predicted that by 1993 an earthquake of magnitude 6 or stronger would strike the Parkfield segment of the San Andreas Fault, where they had calculated a 22-year recurrence interval. No such quake has occurred with only six months to go in the prediction window.

Moreover, though foreshocks are known to occur before about 6% of all Southern California quakes, scientists are unable to determine whether a foreshock has occurred until after the main quake.

In the meantime, research by such scientists as Thomas H. Heaton at the U.S. Geological Survey in Pasadena, James Rice at Harvard and James N. Brune at the University of Nevada has sharply questioned whether the traditional concept provides either an accurate or complete description of how quakes occur.

These researchers postulate that faults may be quite liquid and weak, lubricated by pressurized liquids at depths of four to six miles beneath the surface, where most earthquakes occur.

Brune, researching “stress drops” that result from earthquakes, found evidence that it takes only minor stress changes to trigger temblors. Heaton noted that this should come as no surprise because earthquakes would be much more violent than they are if all the stress was removed and they would generate substantial heat, which has not been detected.

Heaton has also noted in scientific papers that the mechanisms of many quakes cannot be studied because they begin far underground. He contends that this makes the rendering of accurate predictions very difficult.

He and Rice further argue that the differences between large and small quakes in these subterranean zones hinge on mainly unobservable phenomena, such as whether impediments in the pores through which liquids flow may stop a small quake from becoming a large one.

Rice, who has done considerable work in this area, said research at oil drilling sites has frequently revealed highly pressurized liquids deep beneath Earth’s surface.

When these liquids percolate up from very low levels--perhaps even below Earth’s mantle--they may cause wave motion characteristic of an earthquake, Rice said. In this scenario, it would take only relatively minor stress changes along a fault to trigger a quake.

But unanswered questions remain. “There is no good answer as to what builds up the pressure” on the liquids, Rice said. In addition, research in fault zones has proved difficult because of the problem of instability deep in those zones, he said.

“We know now, however, that the eventual amount of slip in earthquakes has only minimal influence over stress . . . and that the reason you are able to have earthquakes at low stress is” an increase in the pressure of water flowing through the spaces, or pores, between rocks.

Traditionally, seismologists have viewed a quake as ripping quickly through the length of the fault zone.

But Heaton, Brune and Rice believe that a quake spreads in a wave or ripple action through pressurized pore liquids, with different parts of the fault moving in quick succession one after another until rocks or other impediments halt the process.

In this scenario, the chance location of a rock could mean the difference between a large or a small quake. And because such deep impediments cannot be examined, they argue, predictions about when a big quake will occur are inherently unreliable.

“We currently record an average of 40 earthquakes per day in Southern California,” Heaton said, “and there is every reason to believe that there are many more that are too small for us to record. Which of these will be a big earthquake? In order to answer this, we need to understand rupture dynamics (better).”

Although some scientists, including some of Heaton’s colleagues at the U.S. Geological Survey, maintain that traditional concepts still explain most earthquakes and that many may be predictable, two of California’s most important earthquakes in the last five years did not conform to the traditional model and have strengthened the arguments of the challengers.

The magnitude 5.9 Whittier Narrows quake of Oct. 1, 1987, and the magnitude 7.1 Loma Prieta quake of Oct. 17, 1989, occurred deep underground and neither left any direct sign of ground rupture on the surface.

The Whittier Narrows quake took place on a subterranean fault that scientists said stretched from Whittier under downtown Los Angeles and Santa Monica and out to sea off the Malibu coast. The fault, they said, could produce a magnitude 6-plus earthquake every 125 to 225 years.

Researchers added that there are probably many other unmapped subterranean faults in the Los Angeles area, some of which might have the potential to give rise to strong quakes.

The Loma Prieta quake was centered in the Santa Cruz mountains on what may have been a strand of the San Andreas Fault. The temblor did not relieve a great deal of stress and what stress relief there was ran perpendicular to the main stress lines of the San Andreas.

Because predictions for various segments of the San Andreas rely on surface indications of previous earthquakes, the more than 10-mile depth of the Loma Prieta earthquake raised the question of how many other such quakes may have occurred, throwing these estimates askew.

Among those sticking with the traditional concepts is Allan Lindh, a seismologist at the U.S. Geological Survey regional office in Menlo Park.

Although Lindh conceded “we’re a very long time from understanding the faults so well we can compute what is happening,” he contends that “the old theory of stress, the old model that energy accumulates and then is released is the best verified model for most earthquakes.”

In many places where the fault intersects the surface, “we can see the deformation of rocks, so we know there is strain,” Lindh added, although he acknowledged that it has been established that stresses are lower than had been believed.

Still, he said, the new theories explain only “anomalies,” departures from the norm that constitute about 20% of earthquakes.

Although the new theorists agree that some quakes conform to the old models, they say their findings that small stress changes can lead to major quakes, the appreciation of the role of pressurized pore liquids, and the belief that impediments under the surface can cause a sharp differentiation between great and small quakes profoundly affects the theory of most earthquakes.

Problems With Predicting Quakes

Recent earthquakes, such as Whittier Narrows and Loma Prieta, have given rise to new theories that make reliable quake predictions in the foreseeable future appear less likely.

While scientists have known for years about surface faults in the Los Angeles area, they have only recently become aware of many subterranean faults in the same area. The subterranean faults were tentatively mapped starting only five years ago, after the Whittier Narrows earthquake. These faults have no surface indications, and there is no information about their past records of rupture that would allow scientists to estimate any intervals between destructive quakes. In the foreseeable future, there is no prospect being able to predict quakes on such faults. New theories that only minor changes in stress--rather than major stress buildups--may cause most quakes also complicate the problems of earthquake prediction.

Impediments in fault zones may halt the growth of small quakes into large ones. If, as some scientists now theorize, quakes are spread through wave, or ripple, action of pressurized liquids instead of representing an instantaneous ripping through the affected fault zone, then the difference between small and large earthquakes may simply be the presence of impediments blocking the spread of liquids. Where the waves spread unimpeded, the quake becomes large. But from the surface, it may be impossible to predict whether a small quake will result in a large one.