Abstract
We present 3-D Vp and Vp/Vs tomographic images for depths down to 22 km along with precise earthquake locations beneath the Sea of Marmara using the seismological data sets collected during the Ocean Bottom Seismometer (OBS) experiments conducted in 2001 and 2006. The unique data set collected by 52 OBSs and augmented by 14 land stations include 3852 P and 2643 S arrival times from 434 earthquakes and 4744 P arrival times from 557 air-gun shot records. From the tomographic depth sections and relocated seismicity, we identify two shear zones which are marked by narrow-width (∼10 km), low Vp (4.3-5.3 km s-1) and high Vp/Vs ratio (∼2.0) anomalies coinciding with the surface trace of the Main Marmara Fault (MMF) and extending near vertically from 8 km depth below the basins into the deeper part of the crust. We consider these shear zones to be the strike-slip deformation signature of the MMF at depth. The western shear zone imaged lies between the Western High and the eastern end of the Central Basin including the 60-km-long segment of the MMF and contains higher seismicity and is visible in the 8-22 km depth range. The eastern shear zone, coinciding with the NW-SE trending surface trace of the MMF, is imaged beneath the Çlnarclk Basin with lower seismicity and tomographic resolution. Furthermore, these shear zones may contain fractures filled with deep crustal fluids allowing partial aseismic creep. Within these shear zones, we also identify small-scale, high Vp (6 km s-1) anomalies with low seismicity indicative of strong patches of the fault at depth that can be considered as asperities. Between these shear zones, our tomographic images clearly show a central zone that has homogenously high Vp (∼6.0-6.5 km s-1) and low Vp/Vs ratio (∼1.7) over a large area including the 50-km-long central segment of the MMF lying between the eastern end of the Central Basin and western end of the Çlnarclk Basin. Thus, we infer that the central MMF is in a locked state based on its high shear strength, low seismicity and a lack of detectable geodetic slip. Furthermore, a large earthquake might not nucleate on the central MMF since it may be strongly locked at the current level of stress loading, and thus it may behave as a seismic barrier or anti-asperity. On the other hand, the central MMF is a good candidate for a supershear rupture because it has nearly uniform elastic properties and low background seismicity indicating the homogeneity of friction and pre-stress across the fault, which are presented as the main characteristics of supershear fault ruptures. From the seismic hazard perspective, we speculate that a large earthquake may be expected to nucleate within either of the imaged shear zones with the possibility of the central MMF acting as a seismic barrier to stop or to join the rupture in a supershear mode. The imaged features have important implications for various aspects of seismic hazard and crustal dynamics in the Sea of Marmara region.
Original language | English |
---|---|
Pages (from-to) | 1172-1187 |
Number of pages | 16 |
Journal | Geophysical Journal International |
Volume | 223 |
Issue number | 2 |
DOIs | |
Publication status | Published - 1 Nov 2020 |
Bibliographical note
Publisher Copyright:© 2020 The Author(s). Published by Oxford University Press on behalf of The Royal Astronomical Society.
Funding
We would like to thank masters, crew, technicians and researchers of the Seismarmara-2001 and Eosmarmara-2006 seismic experiments which were funded by The Scientific and Technological Research Council of Turkey (TÜBITAK) (Project No 102Y114 and 105Y369) in Turkey, INSU,CNRS in France and European Union "SALVADOR" programme of IFM-GEOMAR in Germany. Land station data set were provided by MRC-EMSI and KOERI-BOU. We thank Esen Arpat for general discussions on the geology and tectonics of the Marmara Sea region. We also thank Editor Andrea Morelli, the reviewer Marco Bohnhoff and one anonymous reviewer for their constructive comments. The first author would like to thank Ayşe Kaşlýlar Şişman and Hayrullah Karabulut for their valuable suggestions on the tomographic data set that helped improve the quality of tomographic images and Ekrem Zor for sharing his experience on seismic tomography and help with the use of SIMUL2000 tomography code. Most of the figures were prepared using Generic Mapping Tool (Wessel & Smith 1998).
Funders | Funder number |
---|---|
INSU | |
KOERI-BOU | |
MRC-EMSI | |
Marmara Sea region | |
Leibniz-Institut für Meereswissenschaften | |
Türkiye Bilimsel ve Teknolojik Araştirma Kurumu | 102Y114, 105Y369 |
Centre National de la Recherche Scientifique | |
reviewer Marco Bohnhoff |
Keywords
- Crustal imaging
- Earthquake hazards
- Seismic tomography