Holocene eustatic changes and coastal tectonics in the northeast Mediterranean: implications for models of crustal consumption

Abstract

The African and Eurasian plates are converging at 2.6 cm/a and the Turkey plate is moving west at about 3.0 cm/a. The Aegean is thus a zone of two-dimensional convergence. The boundaries of plates and sub-plates in this area cannot be defined at all simply. Observations of relative changes of sea level were made at 202 ancient harbour sites, some dating back to 2000 b.c., in Greece, Turkey, and Cyprus, and 175 independent estimates of sea level change were made. Measurements of uplifted and submerged recent marine solution notches in Antikythera, Crete, Karpathos, Rhodes, and Cyprus, showed that the islands of the Hellenic Arc are subject to intermittent seismic movements with a periodicity ranging from 157 to 714 a. The archaeological data and the notch data are combined to show that blocks with typical horizontal dimensions of 100 x 50 km at sea level are tilting intermittently as rigid units. The rates of tilt vary from 3.6 to 10.6"/ka. Stable blocks were identified in south Turkey east of Antalya, north Cyprus, and southwest Turkey from Kusadasi to Marmaris. The active areas are: the Peloponnese, where the peninsulae of Argolis and Messenia are subsiding; the central Peloponnese is probably doming or folding, while a slightly uplifted ridge extends east of south through Kythera and Antikythera; a graben structure separates Antikythera from Crete; west Crete is tilting northeast at a rate of 10.6" /ka; eastern Crete is tilting northwards at 10.3" /ka; Karpathos has subsided about 1.0 m maintaining horizontality; Rhodes is folding about a plunging anticlinal axis trending approximately east-west; in Turkey the Izmir-Cesme peninsula is subsiding, as is the coast from Fethiye to Cape Gelidonya; the south coast of Cyprus is subsiding irregularly. These data are used to construct a model of the subduction zone beneath the Hellenic Arc. It is concluded that there is no single planar or slightly curved subducted slab, since the zone bends through 90° south of Crete and again near the Anaximander Mountains. The quasi-linear sections of the subduction zone are interrupted by short tear faults and abrupt changes of strike at intervals of 50-100 km along their length. The resulting narrow fingers of slab tend to break, and are not subducted to great depth. Combination of this model with geophysical data allows construction of a new system of probable plate boundaries. It is suggested that there is no Aegean plate. The north Aegean is an extension of the Turkish plate, while the south Aegean is a region of crustal extension produced by gradual break-up of the southern margin of the Greek-Apulian plate. It is concluded that the Hellenic Arc is not an island arc in the normal sense, but represents a late phase in the destruction of a true island arc and back-arc sea by incipient continent—continent collision.