REVISITING THE SOURCE MECHANISM OF THE AD 365 EARTHQUAKE CRETE, GREECE, AND IMPLICATIONS FOR EASTERN MEDITERRANEAN TECTONICS (Invited Presentation)

A compilation of 32 new and 40 published ¹⁴C dates from a Holocene paleoshoreline encircling the island of Crete is used to test the hypothesis that the shoreline was uplifted as much as 9 m during a single, tsunamigenic earthquake in AD 365. Our modeling of ¹⁴C, fault, and tsunami data collectively support an alternative interpretation that coseismic uplift occurred in two or more earthquakes that ruptured known active normal faults bounding Crete’s southwestern and western coastlines. These contrasting interpretations of the historical earthquake record have a significant bearing on seismic hazards and active deformation in the Mediterranean subduction forearcs. Forearc extension is a hallmark of Mediterranean subduction zones; most active forearc faults are extensional, and little evidence exists for frequent great earthquakes. Nevertheless, the uplifted shoreline has been explained by an M_w 8.3 – 8.5 rupture of a steeply dipping reverse fault that soles into the gently dipping plate boundary beneath Crete (cf., Shaw, et al., 2008). If correct, this represents the most significant historic earthquake in the Mediterranean and one of the largest known coseismic vertical uplift events recorded anywhere.

In contrast, the ¹⁴C results show age distributions at and below the paleoshoreline that span a period coincident with a series of historical and archeologically-documented earthquakes in the first centuries AD. Based on the absence of historically damaging earthquakes after this period, and the reasonable assumption that uplift occurred largely coseismically, we infer that integrated uplift occurred in several events within a few centuries. We identify known active offshore normal faults adjacent to the locus of uplift as potential causative structures. Using visco-elastic dislocation modeling, we show that the rupture of these faults in two < M_w 7.8 earthquakes can reproduce paleoshoreline uplift as well as a single reverse fault event. Tsunami modeling supports this hypothesis, as a normal fault rupture matched historical accounts better than a reverse fault. Our findings favor the interpretation that damaging earthquakes and tsunamis in the Eastern Mediterranean originate on normal faults, a finding consistent with geologic-timescale observations of Hellenic forearc deformation.