The horizontal velocity field in the deforming Aegean Sea region determined from the moment tensors of earthquakes

Abstract

We use the spatial distribution of seismic moment tensors of earthquakes in the Aegean region over the time interval 1909–1983 to recover a continuous horizontal velocity field that describes the overall deformation of the lithosphere at large length scales. The calculated velocity field is dominated by two effects: (1) an E‐W right‐lateral shear of the eastern Aegean, related to motion on the North Anatolian fault becoming distributed as it enters the Aegean; and (2) a N‐S extension, probably related to the sinking of the slab in the Hellenic Trench. The southern part of the central Aegean is found to be moving in a SW direction relative to Europe at a rate of about 30 mm/yr (probably a lower bound, with an error of around ± 10 mm/yr) and rotating clockwise. In the seismogenic upper crust this velocity field is accommodated by right‐lateral strike‐slip faults in the eastern Aegean and by normal faults that rotate clockwise in central Greece. A comparison of paleomagnetic declination data with the expected rotation of rigid elongate inclusions in the velocity field shows in most places an agreement in sense and approximate agreement in rate of rotation. Expected rotation rates of line elements are sometimes too low: probably because our derived velocity field is smoothed and unable to match locally high strain rates. There is only one part of the region where line elements are predicted to rotate in either clockwise or counterclockwise directions, depending on their orientation; this is in western Turkey, which, coincident ally, is the only place where paleomagnetic rotations in both directions have been observed. This coincidence in particular suggests to us that the analogy of rigid elongate inclusions in the velocity field, responding to forces on their bases, may be useful in predicting the senses and approximate rates of rotation of crustal blocks in deforming continental regions. The velocity field we obtain preserves the strike directions of the major faults as directions of zero length change, in spite of considerable smoothing. We use this observation to speculate that the interaction between the upper crust and the rest of the lithosphere beneath it may involve an interplay of effects. On one hand the variation of strength with direction in the crust may control the strike directions of faults that form or become reactivated and may also limit the velocity fields that are allowable. On the other hand, the fault bounded blocks may rotate in the velocity field in response to forces on their bases, which would require the velocity field to change with time if the directions of zero length change are fixed to the blocks.