Gaussian probability estimates for large earthquake occurrence in the Jordan Valley, Dead Sea rift

Abstract

Abstract A catalog including only large earthquakes in the Jordan Valley segment of the Dead Sea rift was compiled using existing recent catalogs for this region. Large earthquakes in the Jordan Valley did not always break the entire plate boundary and, in particular, the last large earthquake (1927) broke only a southern segment of the valley. Thus, the catalog was divided into northern and southern zones (segments). A conservative approach was used whenever the length of the epicentral area was not well constrained. These data were used to estimate Gaussian conditional probabilities for the occurrence of a large (rupturing the plate boundary) earthquake during a 50-year time window, versus the time elapsed from the last large earthquake. Since the historical data, particularly the location and extent of the epicentral areas are not always accurate, the results can be used for gross estimates of the probabilities and generalized conclusions. In general, the analysis show that for the southern Jordan Valley, conditional probability estimates are low mainly due to the short time which elapsed since the last large earthquake relative to the mean repeat time. In the northern Jordan Valley, acceptable probability estimates are high. In addition the probability that a large earthquake would have occurred by 1986 is close to 100%. Such unusually high values most probably mean that some input parameters are erroneous; either the mean repeat time is much larger and the region was the site of much less seismic activity than indicated in the catalog, or, more probably, the last large earthquake is more recent than generally thought. Whatever the explanation, the various computations clearly indicate a higher probability estimate in the northern than in the southern part of the Jordan Valley. The results stress the need for improved studies of the historical seismic record, particularly stressing accurate determination of date and rupture length of the last large earthquake.