Scientists predicting earthquakes such as the 8.8-magnitude tremor that rocked Chile last weekend "won't happen anytime soon," according to one expert, who says research into earthquake patterns has so far yielded little useful data that can be used in forecasting.

For years, scientists have collected information from known fault lines and studied earthquake patterns, to predict where in the world quakes are more likely to occur.

However, that same data has not led scientists to develop a model they can use to predict a quake within days, weeks, months or even years.

"With the current technology that we have at this stage, it won't happen any time soon" John Bellini, a geophysicist at the National Earthquake Information Centre in Golden, Colorado, told CTV.ca in a telephone interview.

"There are lots of measurable statistics and measurable parameters that they can measure for earthquakes and for particular regions, but so far they haven't yielded any repeatable results," he said. "Maybe a lot more study, a lot more instrumentation -- which of course is expensive -- may some day, but not in the near future."

According to the United States Geological Survey, after an earthquake, scientists can calculate the amount of stress that quake has moved onto nearby faults -- and then predict the region's risk of another large tremor.

And if that region has a history of earthquakes, say one every 50 to 100 years, experts can estimate when that quake might occur.

However, compiling that kind of information failed to yield reliable prediction-related data in a massive study that began in California in the 1980s.

In the early '80s, scientists noticed a pattern of 6.0-magnitude earthquakes along a stretch of the San Andreas fault that runs through a small town called Parkfield.

The quakes were spaced out every 20 to 25 years, with the most recent one occurring in 1966. The USGS and the state of California predicted another quake for sometime in the mid-80s and scientists scrambled to set up instruments and record data they hoped would predict the quake.

The quake failed to strike the 1980s. And when one finally hit the region in 2004, none of the information the scientists collected signalled the coming quake: there were no foreshocks, and no electric or water signals.

Other studies have suggested that animal behaviour may predict a coming earthquake. However, the USGS says animals likely only feel smaller shakes moments before a quake, hardly enough time to warn thousands or millions of residents to flee to safer ground.

Meanwhile, as studies fail to establish a model for predicting earthquakes, public officials can only focus on trying to prevent mass deaths and destruction when a quake does strike.

"The money, as far as what happens after an earthquake is located, is being spent on engineering, mitigation and education of the public on how to respond to an earthquake," according to Bellini. "And that's the stuff that's the most useful in saving lives and buildings."

Quake hotspots

While scientists may never be able to accurately predict an earthquake, there are some cities that are virtually famous for them and have a pretty high likelihood of experiencing more in the future.

Few experts would have been surprised by the Chilean quake, given that the strongest quake ever recorded, at 9.5, occurred in the country in 1960.

A study published in the journal Physics of the Earth and Planetary Interiors last year said the Concepcion-Constitucion area "has the potential for an earthquake of magnitude as large as 8-8.5, should it happen in the near future." The study said the region "is very likely a mature seismic gap," meaning its active fault had a build-up of stresses due to little or no seismic activity.

Perhaps one of the most famous earthquake-prone cities is San Francisco, which the USGS says experiences a quake about every 140 years, with some variation, along the Hayward-Rodgers Creek Fault on the eastside of San Francisco Bay.

A forecast on the USGS website estimates there is a "63 per cent probability for one or more magnitude 6.7 or greater earthquakes" in the San Francisco Bay Region before 2036.

The agency says the probability of a large earthquake along the San Andreas Fault in the next 30 years is about 21 per cent. The devastating 7.8-magnitude quake of 1909 occurred along this fault, as did the 6.9 Loma Prieta quake of 1989.

According to Bellini, studies show Los Angeles, which also sits along the San Andreas Fault, has a similar, if not better, chance of experiencing a near 7.0-magnitude quake. "But when that's going to happen is not known," Bellini said.

Other potential hot spots include:

  • Alaska, which the Alaska Seismic Hazards Safety Commission says experiences more quakes than any other region in the U.S. Since 1900, the state has had an average of one magnitude 8 or larger earthquake every 13 years. The second-largest earthquake ever recorded, with a magnitude of 9.2, took place in southern Alaska in 1964. "It is not possible to predict the time and location of the next big earthquake, but the active geology of Alaska guarantees that major damaging earthquakes will continue to occur," the commission says.
  • The Pacific Northwest, including British Columbia, where the Juan de Fuca plate dives beneath the North American plate. Scientists predict that stress that has built up in the Cascadia subduction zone, where the two plates meet and which has been dormant for 300 years, will eventually cause the plates to slip, resulting in a massive quake. The quake that hit the region in 1700, a magnitude 9, spurred a tsunami that sent waves reaching as high as 12 metres crashing onto the coastline.
  • Istanbul, which lies on the North Anatolian Fault system and has been rocked by an earthquake twelve times over the last 15 centuries. A study published in the journal Science found a 62 per cent probability of another strong quake before 2030.
  • Manila, the Philippines. A study published in the Bulletin of Seismological Society of America last year found that the Marikina Valley fault system on the northeast edge of Manila "poses a greater seismic hazard than previously thought." The research suggested a quake of between 6 and 7 is likely, but did not suggest when.
  • Japan, which has been preparing for the next big earthquake in the Tokai region since 1976. According to the Tokai Earthquake Theory, the Philippine Sea Plate has been levering itself under the Continental Eurasia Plate, pulling down its tip. When the pressure reaches its limit, the Eurasia Plate will bounce back, causing an earthquake. The last Tokai quake, the Ansei Tokai Earthquake, occurred in 1854, and experts believe the next big one is on its way. The region expected to be hit hardest is about 160 kilometres southwest of Tokyo.