The conventional wisdom which says if a country like Israel has not had a major earthquake for a long time it is likely to have one soon, is wrong, according to Bar-Ilan University physics Professor Shlomo Havlin, who has detected a â€œclusteringâ€ phenomenon in small and major earthquakes around the world. Havlinâ€™s new research, which suggests that earthquakes have what his team calls â€œmemoryâ€ in their recurrence time, is published in the November 11 issue of the prestigious journal Physical Review Letters of the American Physical Society. â€œThis is the first time that memory has been found in the occurrence of earthquakes,â€ said Havlin, a former dean of BIUâ€™s faculty of exact sciences. The last devastating earthquake here occurred in 1927 in the Dead Sea area, killing about 300 people at a time when the land was sparsely populated. Scientists and government officials have been worrying that Israel is thus due to have another â€œBig Oneâ€ at any time within the next few decades. But Havlin, working in collaboration with Prof. Armin Bunde of the Justus-Liebig University in Giessen, Germany, and Bar-Ilan University graduate student Valerie Livina, borrowed the scaling approach from physics to develop a mathematical function that characterizes earthquakes of a wide range of magnitudes. This makes it possible, he told The Jerusalem Post, to learn from smaller magnitude earthquakes about larger ones. Havlin and his team concluded that the recurrence of earthquakes is strongly dependent on the recurrence times of previous earthquakes. Thus two earthquakes separated by a short recurrence time, for example, are likely to be followed by a third earthquake that recurs close in time, whereas two earthquakes separated by a large recurrence time are likely to be followed by a third earthquake striking further along. â€œI am not talking about aftershocks of an earthquake, but about completely different earthquakes that come in clusters,â€ Havlin explained. â€œWe donâ€™t claim that we can predict when the next earthquake will occur in any one place, but maybe we can improve existing prediction somewhat. â€œWe have shown that the recurrence of earthquakes it not random, as what was believed, but goes in a pattern. If there are places where the frequency of earthquakes was below average, then future ones will also tend to be below the average frequency,â€ he said, adding there is no difference in the phenomenon between large and small tremors. The memory effect provides a clue to understanding the observed clustering of earthquakes, like those which have recently occurred in Asia, and also explains the relatively long delay in the major earthquakes which are expected to hit Tokyo and San Francisco. The span of time since the last major earthquakes in those cities is significantly greater on average than the time span between earlier major earthquakes. â€œWhat we are seeing now in Asia could be an indication of a clustering of earthquakes,â€ Havlin said. At a recent Science and Technology Committee session called on earthquake prediction, scientists at the national Geological Institute in Jerusalem stated that current scientific knowhow does not allow a prediction of when a powerful earthquake will occur that would allow residents and government to prepare for it. There is a â€œgeneral formulaâ€ of the behavior of earthquakes in location and time, but calculating chances for a devastating quake is problematic, they said. A flow of radon gas before weak earthquakes in the Jordan Rift has been detected, but that discovery, they said, does not help predict major ones. Due to the uncertainty, the Geological Institute recommended improving building standards and strengthening existing structures to be ready for earthquakes, whenever they occur.