PALEOSEISMOLOGY OF THE NORTH ANATOLIAN FAULT, A PERSPECTIVE

Seismic risks of the North Anatolian fault became an urgent and important issue after the 1999 Kocaeli (Izmit) earthquake. The hypothesis of heightened risks in Istanbul based on the analyses of historic data and stress transfer model, for example, was a significant application of paleoseismology. However, for the major part of the North Anatolian fault, geological investigations on past earthquakes have not presented enough unique, or useful parameters to assess the risks. This paper aims to compile and review current geological data on the fault behavior and to give a perspective for the future studies. (1) The most recent event in the east of 1939 segment is supposed to be 1784, but there is no geologic information in this area. (2) The penultimate events in and around Marmara area occurred in 18th century A.D. Historic records strongly support this, but geologic data have not attained enough time-resolution to define the timing of the penultimate events. The great 1668 earthquake(s) ruptured 1939 (not entire?), 1942, 1943, 1944 and possibly 1967 segments of the last events. However, the evidence from trenches on the segments does not consistently support the occurrence of the 1668 event. (3) Recurrence intervals are 200 to 250 years in average for three past events in Marmara area, but both short (200 to 300) and long (500 to 700) recurrence time hypotheses are proposed for the segments in the east. Also, both regular and irregular recurrence hypotheses are possible from those data. (4) The slip during the last events need to be carefully re-examined together with precise mapping of the fault ruptures. (5) There is very limited data on the slips during penultimate events, or the cumulative slip in definite span of time. Then, we do not have well constrained slip rate for thee most part of the fault. High-resolution careful mapping of offset features by the last and older events are critical.