A new study published by researchers from Bar Ilan University and Germany revealed "memory" patterns in the dates of earthquakes' appearances
Do earthquakes have a memory, or is it perhaps a random phenomenon that occurs without warning at any time? Is it possible to derive insights from the devastating earthquakes of the past year that will improve the ability to warn of future earthquakes?
A new study conducted by Prof. Shlomo Havlin from the Physics Department at Bar Ilan University and his colleagues from Israel and Germany answers this question in the affirmative. The research reveals that earthquakes create statistical patterns, with the help of which it will be possible to improve the predictability of their recurrence.
Prof. Heblin's research, which was carried out in collaboration with Prof. Armin Bunde from the Justus Leibig University from the city of Giessen in Germany, and the doctoral student from Bar Ilan Valerie Levina, was published (11/11/2005) in the prestigious scientific journal Physical Review Letters ), published by the American Physical Society.
Prof. Heblin and his team developed, using the scaling theory accepted in statistical physics, a mathematical function that tries to describe the statistics of the time intervals between earthquakes of similar intensities. The calibration theory makes it possible to extrapolate from low-intensity vibration characteristics for which there are good statistics, to high-intensity vibrations for which there is no sufficient data for statistical analysis. The researchers analyzed data files of earthquakes that occurred in the last decades around the world. The findings reveal that the time interval until the next earthquake in a certain place depends on the time interval between the previous earthquakes of similar intensity and in the same place. The memory is expressed in the fact that if there is a shorter-than-average time interval between earthquakes in a certain area, there is a high chance that the next earthquake will also occur within a shorter-than-average period of time. And if the time interval between two earthquakes that occurred in a certain place is greater than the average, the time interval until the next earthquake will also be greater than the average.
This memory effect may help to understand the appearance of clusters of strong earthquakes (Clustering) like those that occurred in Asia, and even explains the relatively long wait for the next big earthquake in Tokyo and San Francisco. The time interval since the last major earthquakes in each of these cities is significantly greater than the average time intervals between the previous major earthquakes there.
Prof. Heblin and his team analyzed the historical statistics of repeated appearances of earthquakes above certain intensities (from 2 to 5.5 on the Richter scale). They used data collected from records of seismic activity at active sites around the world, including northern and southern California, Japan and New Zealand.
"This is the first time that memory has been discovered in the appearance of earthquakes," says Prof. Havlin, former dean of the Faculty of Exact Sciences at Bar Ilan University. "What we are seeing now in Asia could be a sign of the occurrence of a cluster of large earthquakes," explains Havlin, who added that his team's research may be used by both scientists and seismologists.
