Seismologists Grumble About `Quack' Quake Predictions

The residents of New Madrid, Mo., were ready. They bought their emergency supplies and reinforced their windows. But no great earthquake violently rocked the area on December 3, as New Mexico business consultant Iben Browning had predicted. The ground didn't even quiver. Maybe the earth didn't tremble, but Browning's prediction did set off some rumblings in the scientific community. "We [seismologists] are very angry with the news media," says Max Wyss, a geology professor at the University of Colorado, Boulder, and a researcher at that school's Cooperative Institute for Research in Environmental Sciences (CIRES). "I even wrote a letter to [CBS News anchor and managing editor] Dan Rather about it." What angered Wyss and other seismologists was not so much the fact that Browning made such a prediction, but that those in the media did not bother to challenge or check the scientific validity of the forecast. "Reporters have to double-check their sources," says Wyss. Both Wyss and Timothy Long, a professor in the Georgia Institute of Technology's School of Geophysical Sciences in Atlanta, cite the highly publicized, unsound prediction as the type of incident that could leave a stigma on seismology in particular and science in general. "In the short term," says Long, "it [the prediction] made people aware that quakes do exist. In the long term, it can leave a mark on the science [of earthquake prediction]." Such a "bad reputation" worries seismologists, a group of researchers who range from the top scientist at the National Center for Earthquake Engineering Research in Buffalo to the geology professor at a small, liberal arts college in Philadelphia who built his seismograph from scratch. Publicity of this type, they say, could mar the general public's image of the field and could even sway the opinion of those who hold the purse strings--a possibility that is especially disheartening at a time when all scientists are lamenting the lack of sufficient funds to support their work. "We try to educate the public about the process of earthquake prediction," says Wyss. "The first thing we have to do is flag a quack prediction.... That type of sensationalism does damage to science." That, adds Wyss, could influence those in government who allocate funds: "They'll say, `We don't want to fund stuff like that.'" Another factor that has adversely affected the image of seismology is the difference between seismologists' and the public's perception of what constitutes a successful prediction. Says Wyss, "The general public would like accuracy of a day or week." But seismologists, he says, are "pleased with one year." Wyss and his colleagues have been monitoring a section of the San Andreas fault near Parkfield, Calif. This region of the fault, which lies at the midpoint between Los Angeles and San Francisco, has averaged one moderate earthquake (magnitude 5.5 to 6 on the Richter scale) every 22 years since 1857. Armed with this knowledge, and almost two decades' worth of seismological data collected by the U.S. Geological Survey from this region, Wyss and his associates have estimated that the next moderate earthquake to strike the region may do so in March 1991, give or take a year. The team has based its prediction on the analysis of a number of precursory phenomena, such as an increase in the average depth of small earthquakes in the region, fluctuations in the average magnitude of the earthquakes, seismic quiescence (a drop in the number of small tremors), and a decrease in the rate of crustal deformation. Although Wyss believes strongly in his data, he's not about to be lulled into a false sense of security. "I'm confident that the phenomena I am reporting do exist," he says. "But that, by no means, indicates that an earthquake is definitely coming. We'll always have to live with the notion of false alarms." False alarms, however, do not diminish the frequency of occurrence of earthquakes. All throughout the geologic column, scientists have found evidence of seismic activity. And for as long as there have been people on the earth, there have been numerous attempts to explain this activity. Everything from giant, subterranean catfish to avenging angels have been blamed for the destructive rumblings in the earth. It wasn't until the mid-1800s that a more geological explanation was sought. Robert Mallet, an Irish engineer, proposed that earthquakes took place within the earth's crust when the bedrock ruptured in response to accumulated elastic strain. In 1929, when Alfred Wegner introduced his theory of continental drift (the groundwork for the modern theory of plate tectonics), he gave geologists a source for these stresses. As the jigsaw puzzle pieces of the earth's crust grind past, slide beneath, pull apart from, and collide with each other, they give rise to mountain ranges, volcanoes, oceanic basins, and earthquakes. One of the biggest boosts to seismology came about somewhat by accident. In the 1950s, the World Wide Standardized Seismograph Network (WWSSN) was proposed as a means to monitor underground nuclear explosions. The system, completed in 1967, easily picked up large blasts. Unfortunately, it could not pick up the smaller blasts associated with nuclear weapons tests. These smaller explosions were masked by the earth's constant seismic activity. As a weapons monitoring system, WWSSN failed. As a source of seismic information, however, it flourished. An earthquake occurring anywhere in the world is automatically recorded by enough stations to pinpoint its focus (place of origin within the crust), epicenter (point on the surface directly above the focus), and magnitude (size or energy released by the quake). With all this information, then, why is it still difficult for seismologists to accurately predict where and when an earthquake is likely to strike? First of all, says Wyss, there are "too many parameters"--such as tectonic setting, rock type, water content, and previous seismicity--to try to fit into a truly accurate model. In addition, past records for many regions are incomplete. "We have very few case histories," says Wyss, referring to information that allows seismologists to glean predictable patterns in seismic activity. "And," he adds, "earthquakes happen where they happen and when they happen. They don't happen in our laboratories." While it's true that earthquakes don't occur in laboratories (unless the lab is strategically built on top of an active fault), it is in the lab where some researchers are attempting to understand the mechanisms that trigger earthquakes. One type of earthquake that has baffled seismologists and geophysicists for some time is that which takes place deep within the crust, at depths greater than 70 kilometers. These earthquakes occur in regions like Japan, where oceanic crust is subducted beneath the lighter continental crust. At these subduction zones, crustal material enters the next layer in the earth, known as the mantle. Scientists have argued that at these depths, the surrounding pressure and temperatures are great enough to prevent the crustal rocks from fracturing. Instead, they should just deform in a ductile fashion, like modeling clay under pressure. Geophysicists at the University of California, Davis, have conducted experiments in order to come up with a reasonable explanation for such earthquakes. In their experiments, geology professor Harry W. Green II and his doctoral student subjected the low-pressure analog of the mineral olivine (a major component of the mantle) to increasing temperatures and pressures. Under such conditions, olivine undergoes a phase change; that is, it changes to spinel, a mineral that is chemically identical to olivine but has a different crystalline structure. Green's group found that within a narrow range of temperature, lenses of spinel line up and link, forming zones of weakness along which quakes can occur. Long also is working on a model to explain the occurrence of major earthquakes. Long's research, however, is focused on the strong quakes that strike within the interior of a plate--as opposed to those that happen where two plates come together. Several of this major type of earthquake have already struck the eastern United States, from Massachusetts to South Carolina. In fact, the three largest earthquakes known in all of North America rocked the New Madrid region back in 1811 and 1812. These powerful earthquakes changed the course of the Mississippi River and rang church bells in places as far away as Boston. The prevailing geologic concept of the New Madrid region, says Long, is that the area is underlain by an ancient rift that would provide a suitable zone of weakness for seismic activity. "Maybe," says Long, "we should take the opposite viewpoint. Suppose the weaknesses [in the crust] don't already exist." Long's model looks at normal crust and tries to identify the mechanisms (such as changes in temperature due to intruding molten rock or changes in fluid content) that could weaken it, thus leaving it open to fail under stresses that already exist. Such a model could explain the earthquakes that have occurred in tectonically quiet regions. And if the model works, then, in essence, no place on earth is truly earthquake free. But, cautions Long, no single model can accurately describe every earthquake that occurs. "Different types of earthquakes exist--New Madrid; Piedmont, Calif.," he says. "Each type will have to be treated individually. Careful monitoring of minor seismicity and a careful understanding of the minor stresses is needed." Such an understanding, however, will come at a price. "We need a lot more instruments," says Wyss. "And better funding for scientists"--funding that would enable seismologists to work with the information that has already been collected. "There are a lot of data waiting [for interpretation]," he adds. Although the December 3 fiasco could restrict the flow of needed funds, Wyss points out that quack predictions and media sensationalism aren't the only problems seismologists face. The scientists themselves, he says, aren't totally blameless: "A lot of poor work has been done. People want quick results. They [the public] want to see remedies and solutions." Quake prediction research, he says, is going slowly. "When do you say we've got a handle on it?" he asks rhetorically. "There will always be false alarms and surprises."

Seismologists Grumble About `Quack' Quake Predictions

Interested in reading more?

Become a Member of