Abstract
The Iranian plateau accommodates the 35 mm/yr convergence rate between the Eurasian and Arabian plates by strike-slip and reverse faults with relatively low slip rates in a zone 1000 km across. Although these faults have only locally been the subject of paleoseismological studies, a rich historical and archeological record spans several thousand years, long enough to establish recurrence intervals of 1000 to 5000 yr on individual fault segments. Several clusters of earthquakes provide evidence of interaction among reverse and strike-slip faults, probably due to adjacent faults being loaded by individual earthquakes. The Dasht-e-Bayaz sequence of 1936 to 1997 includes earthquakes on left-lateral, right-lateral, and reverse faults. The Neyshabur sequence of four earthquakes between 1209 and 1405 respected the segment boundary between the Neyshabur and Binalud reverse fault systems. The two pairs of earthquakes may have ruptured different faults in each segment, similar to the 1971 and 1994 San Fernando, California, earthquakes. The 1978 Tabas reverse-fault earthquake was preceded by the 1968 Ferdows earthquake, part of the Dasht-e-Bayaz sequence. The North Tabriz fault system ruptured from southeast to northwest in three earthquakes from 1721 to 1786; a previous cluster may have struck this region in 855 to 958. The Mosha fault north of Tehran ruptured in three earthquakes in 958, 1665, and 1830. Five large earthquakes struck the Tehran region from 743 to 1177, but only two that large have struck the area since 1177. Other earthquakes occurred in pairs in the Talesh Mountains near the Caspian Sea (1863, 1896), the Iran-Turkey border (1840, 1843), and the Nayband-Gowk fault system (both in 1981). Other historical events did not occur as parts of sequences.
The historic seismic moment release in Iran accounts for only a small part of the plate convergence rate, which may be due to aseismic slip or to the Iranian historical record, long as it is, being too short to sample long-term deformation across the plateau. No historic earthquakes of M 
8 have struck Iran. However, several long, straight strike-slip faults (Doruneh, West Neh, East Neh, and Nayband) have not sustained large historical earthquakes, raising the possibility that these long faults could produce earthquakes of M 8, thereby removing at least part of the apparent slip deficit.
An increased understanding of Iran's seismic hazard could be obtained by an extensive paleoseismology program and space-geodetic arrays, supplementing the abundant historical and archaeological record.