Intraclast breccias in laminated sequences reviewed: Recorders of paleo-earthquakes

Observations of intraclast breccia layers in the Dead Sea basin, formerly termed "mixed layers," provide an exceptionally long and detailed record of past earthquakes and defi ne a frontier of paleoseismic research. Multiple studies of these seismites have advanced our understanding of the earthquake history of the Dead Sea and of the processes that form the intraclast breccias. In this paper, we describe a systematic study of intraclast breccia layers in laminated sequences. The relationship of intraclast breccia layers to intraformational fault scarps has motivated the investigation of these seismites. Geophysical evidence shows that the faults extend into the subsurface, supporting their potential association with strong earthquakes. We defi ne fi eld criteria for the recognition of intraclast breccias, focusing on features diagnostic of a seismic origin. The fi eld criteria stem from our understanding of the mechanisms of breccia formation, which include ground acceleration, shearing, liquefaction, water escape, fl uidization, and resuspension of the originally laminated mud. Comparison between a dated record of breccia layer and the record of historical earthquakes provides an independent test for a seismic origin. The historical dating is signifi cantly more precise and accurate than the radiocarbon dating of breccia layers. Yet, assuming that the lamination of the sediments shows an annual cycle, the precision of counting laminae may approach the precision of the historical record. A similar accuracy is then expected for the intervals between earthquakes. We review our work based on counting laminae representing the historical period, mutually corroborating the seismic origin and the annual lamination. The correlation of documented historical earthquakes with individual breccia layers provides quantitative estimates for the threshold of ground motion for breccia formation in terms of earthquake magnitude and epicentral distance. The investigation of breccia layers and the associated historical earthquakes has underscored cases in which a breccia layer represents a pair of earthquakes. We consider the resolution of individual events in records of breccia layers. A thick breccia layer can account for multiple events, biasing the paleoseismic record. The resolution of an interseismic time interval is no better than the ratio between the thickness of a breccia layer and the rate of sedimentation. We use revised age data for the Lisan Formation and reassess temporal clustering of earthquakes during the late Pleistocene. The variation of recurrence interval corroborates signifi cant clustering. During periods of clustered earthquakes, of order of 1000-5000 yr, the interseismic interval becomes short, and the resolution diminishes, so the peak rate of recurrence may be underestimated. Recurrence intervals inferred from the Dead Sea record of Holocene breccia layers do not feature the extreme variation encountered in the late Pleistocene record. Yet the Holocene record shows marked transitions between periods, each with relatively uniform recurrence interval. Two of the transitions are contemporaneous with transitions in the recurrence intervals of the Anatolian faults, implying broad-scale elastic coupling.