Vertical velocity fields along the Eastern Mediterranean coast as revealed by late Holocene sea-level markers

Vertical movements of the solid Earth surface reflect crustal deformation and deep mantle related phenomena. For Holocene times, coastlines displaced from the present mean sea-level are often used together with past relative sea-level (RSL) prediction models to decipher the vertical deformational field. Along the coastline from southwest Turkey eastward to Israel and Cyprus, field data that constrain Holocene vertical movements are already published, leaving a gap only along the Mediterranean coast of the Central Anatolian Plateau (CAP). Based on new field observations between Alanya and Adana (Mersin, southern Turkey), together with AMS 14C dating, we fill that gap, allowing for the construction of a continuous overview of Ho-locene vertical differential movements along the Eastern Mediterranean coast. We apply the most recent Glacial Isostatic Adjustment (GIA) models to correct for the glacio-hydro isostatic component of the RSL. Different so-lutions from the ICE-6G(VM5a) and ICE-7G(VM7) models (developed by W.R. Peltier and co-workers at the Toronto University), and a GIA model developed by K. Lambeck and collaborators at the Australian National University, have been applied to 200 middle-to-late Holocene RSL markers. Starting from southwest Turkey, we find subsidence between-0.9 mm/yr and -2.3 mm/yr, corroborating estimates from previous studies. Velocities from the new markers along the CAP Mediterranean coast are posi-tive, ranging between 0.9 and 1.5 mm/yr. These two first blocks are separated by a sharp velocity jump, occurring along the Isparta Angle Fault System one. Such high vertical velocities for the CAP southern margin were predicted by recently published papers that report a rapid uplift phase that peaked during themiddle to late Pleistocene. Moving to the east, velocities are also positive, from 0.2 to 0.6 mm/yr along the coast between the Hatay Gulf and southern Lebanon. The highly variable velocity along the Lebanese sector is likely due to co -seismic deformation along the Lebanese Restraining Bend (LRB) faults. To the south, in contrast, the Israeli coast shows stability, according to some unique archaeological RSL markers named piscinae, whereas other markers indicate slow subsidence (-0.2 mm/yr on average). Hence, another velocity jump of at least 0.5 mm/yr is recognizable between Israel and Lebanon. This jump is probably associated with mapped, active tectonic structures. In Northern Cyprus, the only Holocene sea-level marker confirms the near zero vertical velocity values already obtained for the MIS 5e marine terrace. Therefore, a vertical velocity jump occurs between stable Northern Cyprus and the uplifting CAP southern margin, although they occur on the same overriding plate of the Eastern Mediterranean subduction system. High-angle normal faults at the northern margin of the Adana-Cilicia Basin could explain these strongly distinct late Holocene vertical velocity fields. These results depict a complex framework of independently moving crustal blocks, with kinematic separation along well-known regional fault zones. The drivers of the block movements could be related either to regional tectonics, as it is probably the case for the LRB coast, or to mantle dynamics, such as for the uplifting Turkish sector, where deeper processes should be considered.