1,1,1,2,2,3- 1Alma Mater Studiorum - Università di Bologna, Department of Physics and Astronomy "Augusto Righi", Bologna, Italy (cesare.angeli2@unibo.it, alberto.armigliato@unibo.it, martina.zanetti8@unibo.it, filippo.zaniboni@unibo.it)
- 2Bogazici University, Kandilli Observatory & ERI, İstanbul, Türkiye (didem.samut@bogazici.edu.tr, fatih.turhan@bogazici.edu.tr, ozeln@bogazici.edu.tr)
- 3METU, Department of Civil Engineering, Coastal and Ocean Engineering Division, Türkiye (yalciner@metu.edu.tr)
The Sea of Marmara represents one of the most critical seismotectonic regions worldwide, as it hosts the offshore segments of the North Anatolian Fault Zone (NAFZ), a major right-lateral strike-slip fault system [1, 2]. This fault system has generated numerous destructive earthquakes and constitutes a major seismic hazard for Istanbul, a megacity with a population exceeding 16 million. Most recently, on 23rd April 2025, a Mw 6.3 earthquake ruptured a segment of the NAFZ beneath the Sea of Marmara, producing strong to severe ground shaking across many coastal settlements around the basin. According to the KOERI rapid response report, a potential for light structural damage has been identified, particularly in the districts of Silivri, Büyükçekmece, Beylikdüzü, Avcılar, and Küçükçekmece, corresponding to approximately 0.8 per thousand of Istanbul’s total building inventory. Being located underwater, the earthquake also generated a small tsunami that was observed in many tide gauges (TG) along the coasts of the Marmara Sea.
In this work, we present a thorough analysis of the available sea level data from the day of the event. To start, available TG signals from that day are analysed using modern Time-Frequency (TF) techniques, namely the Fast Iterative Filtering (FIF) [3] that produces a data-driven decomposition of a given signal and the IMFogram [4], to obtain their TF representation. From here, we are able to determine arrival times, amplitude and main periods of the tsunami.
After the TF analysis, we simulate the tsunami through numerical modelling with different fault models using the JAGURS [5] software. First, we consider the case of faults with homogeneous slip distribution that we obtained from available Centroid Moment Tensor (CMT) solutions using scaling laws. Then, we propose a distributed-slip fault model. Such model has been obtained with a grid search like method and is composed of two uniform slip patches: a long shallow section with slip of around 1 m and a deeper and more compact one with slip around 0.66 m. This composite reproduces experimental time series with minimal error and its scalar seismic moment agrees with solutions found in the published literature [6].
At last, the possible presence of a secondary tsunami source is discussed, in order to explain an anomalous signal observed in Armutlu. Arguments in favour of a possible submarine landslide are presented and discussed.
[1] Barka, A. A. (1992). Annales tectonicae (Vol. 6, pp. 164-195).
[2] Şengör, A. M. et al. (2005). Annu. Rev. Earth Planet. Sci., 33(1), 37-112.
[3] Cicone, A., & Zhou, H. (2021). Numerische Mathematik, 147(1), 1-28.
[4] Cicone, A., et al. (2024). Applied and Computational Harmonic Analysis, 71, 101634.
[5] Baba, T., et al. (2015). Pure and Applied Geophysics, 172(12), 3455-3472.
[6] Eken, T., et al. (2025). Journal of Seismology, 1-29.
How to cite: Angeli, C., Armigliato, A., Zanetti, M., Zaniboni, F., Cambaz, M. D., Turhan, F., Meral Özel, N., and Yalciner, A. C.: Tsunami Generation by the Moderate 23rd April 2025 Earthquake occurred in the Marmara Sea, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7949, https://doi.org/10.5194/egusphere-egu26-7949, 2026.
