The study's authors, who work at Columbia University's Lamont-Doherty Observatory, acknowledge that the biggest earthquakes
- in the 6 or 7 magnitude range - are rare in the New York City region. They say a quake of magnitude 7 probably comes about every 3,400 years.
But they note that no one knows when the last one hit, and because of the population density and the concentration of buildings and financial assets, many lives and hundreds of billions of dollars are at risk.
The metropolitan area does not have a single great fault like the San Andreas fault in California, said Leonardo Seeber, co-author of the study.
"Instead of having a single major fault or a few major faults, we tend to have a lot of very minor and sort of subtle faults," he said. "It's a family of faults, and that can contribute to the severity of an earthquake."
John Ebel, director of seismology at Boston College's Weston Observatory, said he agreed with the study's finding that small faults can contribute to large earthquakes. "A quake can jump from one fault to another," he said.
The study, published in the Bulletin of the Seismological Society of America, analyzed 383 known earthquakes over the past 330 years in or near New York City. The biggest were three that reached magnitude 5 in 1737, 1783 and 1884.
Data on earthquakes since the early 1970s, when Lamont deployed dozens of new detectors, enabled the authors to see patterns from smaller quakes, including the magnitude 4.1 quake that was centered on Ardsley, in Westchester County, in 1985.
The report inferred from the data that there is a seismic zone, previously undetected, running west from the southwest tip of Connecticut and intersecting with the large, well-known Ramapo fault near Indian Point.
Lynn Sykes, the lead author, said the finding means the danger of a big quake near the nuclear plants is greater that had been thought.