GNGTS 2013 - Atti del 32° Convegno Nazionale

Stein et al. (1997) showed the earthquake-induced Coulomb stress changes on adjacent fault segments. Application of this technique to evaluate the effect of the recent Kocaeli earthquake (1999, M w =7.4) on neighboring faults, shows an area of increased stress to the east, including the Düzce fault which ruptured just after the Kocaeli earthquake (12 November, 1999, M w =7.2). To the west, both the 80 km-long Yalova segment, southeast of Istanbul, and the Northern Boundary fault, immediately south of Istanbul, may be close to failure (Hubert-Ferrari et al. , 2000). In spite of active morphological expressions along the middle strands from the Mudurnu valley (from the 1967, M w =7.2, earthquake) to the Aegean Sea, there are only a few possible earthquakes for the last 1000 years which can be associated with this strands. Thus, almost the entire middle strands can be considered as a potential site for large earthquakes. Together with that during the last two centuries, the greatest part of the southern strands has been ruptured as a result of large earthquakes. However, two possible seismic gaps still exist; the Pazarkoy- Edremit gap (60 km long with a double bend geometry separated from the 1953, M w =7.1 rupture segment by about 12 km wide restraining stopover), and the Yenisehir gap (has a pull-apart geometry). This branch also shows a larger cluster of seismicity near the fourth biggest city of Turkey, Bursa. The area demonstrates several typical aspects of the regional seismotectonic activity and has been subjected to several large earthquakes in the last centuries (28/02/1855, M w =7.4 and 11/04/1855, M w =6.8). Besides very high earthquake hazard, the earthquake risk in those cities has been increasing steadily due to overcrowding, improper land- use planning and construction, inad- equate infrastructure and services, and environmental degradation. Be- cause of the real earthquake threat in the Marmara region, the need for seismic hazard studies has become progressively more important for earthquake engineering applications. Recently, seismic hazard in Istanbul has been estimated using probabilis- tic methods (Atakan et al. , 2002; Er- dik et al. , 2004). Fig. 1 – It represents the ratio between the 50 th percentiles PGA seismic hazard for a) Poisson and BPT (%10 in 50 years) b) Poisson and BPT+∆CFF (%10 in 50 years), c) BPT and BPT+ ∆ CFF (%10 in 50 years) calculated using the fault segments only. In red: positive amplifications. In blue: negative amplifications. Black lines represent the surface ruptures of each fault numbered as in Tab. 1. 6 GNGTS 2013 S essione 2.1