Tectonic and hazard implications of a very complex piece of crust: Seismic Imaging the Sea of Marmara, NW Turkey

The Sea of Marmara region in NW Turkey, is an area of complex tectonics and poses a high seismic hazard for the capital city of Istanbul. Detailed passive seismic imaging of the complex structures making up the region will improve our understanding of ongoing tectonic processes and their hazard implications.
The structure in Sea of Marmara is formed of multiple ancient fragments of Crust, and has been shaped by the tectonic processes that formed the continent of Anatolia (Figure 1b). Today the area exhibits complex patterns of deformation driven by present-day processes which interact with these ancient boundaries. From an extensional regime in the neighbouring Aegean, to the transtensional deformation caused by the major strike-slip North Anatolian Fault which bisects the region (Figure 1a), the crust today is being deformed and reshaped by ongoing tectonics. These processes pose significant seismic hazard along active fault zones that sit near large population centres such as Istanbul. The structure of the crust in this complex area is likely to be highly variable over short wavelengths, reflecting both ancient and ongoing tectonic processes. To image and understand the region requires detailed analysis of seismic observations.
This project will make use of a comprehensive continuous seismic dataset of over 150 stations, from both temporary and permanent deployments, including data from the Turkish Disaster and Emergency Management Authority AFAD, Figure 2a. A key focus will be on exploring the best way to compare or combine multiple seismic imaging techniques that are often used in isolation (using global and local earthquakes as well as continuous background ambient noise) to construct the most comprehensive high-resolution images of crustal structure of the region possible. Results will used to gain a better understanding of ancient and ongoing tectonic processes as well as the seismic hazard potential of the region. The impact of using 3D velocity models to accurately locate local earthquakes will be explored to relate seismicity to specific fault segments, while an improved understanding of large-scale crustal structure will feed into geomechanical stress models used to assess hazard potential (e.g. Hergert, et al., 2011).
The candidate will assess regional variations in crustal structure using a range of imaging techniques, including:
• Generating a dataset of surface wave dispersion curves from ambient seismic noise and global events.
• Updating a dataset of P to S converted phases from distant global events (Jenkins et al., 2019).
• Using local seismicity to assess P wave velocity structure with local earthquake tomography
• Jointly inverting data to produce new 3D regional velocity models.
• Generating new measurements of crustal discontinuity structure using reflected phases and ambient noise auto-correlations and exploring how these can be integrated with other datasets.
This project will suit a numerate geologist or (geo-)physicist, and is primarily computational in character. While prior coding experience is not required, candidates must be willing to learn to use, adapt, and extend software written in Python and other programming languages. The candidate will gain expertise in seismic data analysis, the handling of large datasets, inverse theory, and the structure of the Earth’s crust.

Click on an image to expand

Image Captions

Fig1_Marmara.png – Figure 1. From Jenkins et al. (2019). Summarized present-day tectonics of Turkey. Sea of Marmara region in green box. NAF = North Anatolian Fault. (b) Major crustal terranes comprising the Sea of Marmara study region, with major faults in green.
Fig2_marmara.png – Figure 2. a) Dataset of 173 Seismic stations from Jenkins et al. (2019) to be used in this project b) cartoon of the depth sensitivity of surface waves as a function of frequency that leads to dispersive properties c) P to S converted phases generated by global events that are used to analyse crustal discontinuity structure.


Cross correlations of background ambient noise between all station pairs in the network will be used obtain shallow sampling surface wave arrivals (Bensen et al., 2007). Deeper sampling surface waves will be selected from regional earthquakes. The surface wave dispersion characteristics will be analysed using Frequency-Time Analysis and combined in an inversion to produce dispersion curves of shear-wave velocity as a function of depth, sensitive to absolute shear-velocity structure of the crust and upper mantle. Dispersion data provides smoothed images of the true velocity, and lacks sensitivity to sharp boundaries (such as the Moho). To combat this a pre-existing dataset of receiver functions (Jenkins et al., 2019) generated from Ps converted phases sensitive to sharp interfaces will be updated and combined with dispersion data in a joint inversion (Gilligan et al., 2015) to produce a new model of 3D seismic shear-velocity variation across the region. Local seismicity in published catalogues (e.g. Wollin et al., 2018) will used as inputs for Local Earthquake Tomography to also consider compressional-velocity structure.
Complementary analysis of converted underside releftions (SsPmP phases) using Virtual Deep Seismic Sounding (Thompson, Rawlinson & Tkalčić, 2019), and ambient noise auto-correlations (Taylor et al., 2016) will be used to improve detailed mapping of crustal thickness variation and internal discontinuity structure. These observations will be compared to the 3D seismic velocity model to assess reliability.
Seismic results will be interpreted in terms of the current understanding of the crustal fragments forming the region and ongoing deformation processes. Velocity models will be used to relocate local Earthquakes to assess the impact of a 3D velocity model in calculating accurate Earthquake locations. Assessment of Moho topography and depth extent of deformation along active fault zones will be used to provide updated inputs for geomechanical stress models.

Project Timeline

Year 1

Literature review, training in seismic data processing techniques, data collection of regional Earthquake surface wave data and FTAN analysis of this and pre-existing ambient noise surface wave data. Updating of Ps receiver function dataset.

Year 2

Joint inversion of ambient noise dispersion, surface wave dispersion and RF data to produce 3D seismic velocity model. The work from Years 1 & 2 should lead to at least one publication.

Year 3

Data collection and analysis (virtual deep seismic sounding (VDSS) and ambient noise auto-correlation) to map crustal discontinuity structure. Assessment of impact of 3D velocity model in accurate Earthquake location and use for rupture modelling working with researchers in GFZ Potsdam.

Year 3.5

Completion of manuscripts for publication and thesis writing

& Skills

The student will become part of the Geophysics and Geodynamics Research Groups at Durham, and the Seismology group at the British Geological survey.
The student will receive training in processing, analysing and modelling seismic data as well as associated essential skills (programming, code development, and usage of high-performance computing systems). Training in a wider range of important skills (e.g. presentation skills, paper/thesis writing) will be provided by the Department of Earth Sciences at Durham University, and the student will also benefit from cross-disciplinary training provided as part of the IAPETUS2 DTP.
The student will have opportunities to work with other partners in the UK and internationally and will attend national and international scientific meetings to present results. We aim to see all students publish at least two papers in leading scientific journals during their PhD. Upon completion, the student will be well equipped for a career in academia or in a range of industries.

References & further reading

Bensen, et al., (2007). Processing seismic ambient noise data to obtain reliable broad-band surface wave dispersion measurements. Geophysical Journal International, 169(3), 1239-1260.
Gilligan, et al., (2015). The crustal structure of the western Himalayas and Tibet. Journal of Geophysical Research: Solid Earth, 120(5), 3946-3964.
Hergert, et al., (2011). Geomechanical model of the Marmara Sea region—I. 3-D contemporary kinematics. Geophysical Journal International, 185(3), 1073-1089.
Jenkins, et al.,. (2020). Crustal thickness variation across the Sea of Marmara region, NW Turkey: A reflection of modern and ancient tectonic processes. Tectonics, 39(7),
Taylor, Rost., & Houseman. (2016). Crustal imaging across the North Anatolian Fault Zone from the autocorrelation of ambient seismic noise. Geophysical Research Letters, 43(6)
Thompson, Rawlinson & Tkalčić (2019). Testing the limits of virtual deep seismic sounding via new crustal thickness estimates of the Australian continent. Geophysical Journal International, 218(2), 787-800.
Wollin, C., Bohnhoff, M., Martínez-Garzón, P., Küperkoch, L., & Raub, C. (2018). A unified earthquake catalogue for the Sea of Marmara Region, Turkey, based on automatized phase picking and travel-time inversion: Seismotectonic implications. Tectonophysics, 747, 416-444.

Apply Now