Comments on “An investigation of seismicity for western Anatolia” by Sayıl and Osmanşahin (2008)

Abstract

Numerous local or regional seismic hazard studies have been made to estimate the seismic hazard in and around Turkey using the statistical processing of historical and instrumental earthquake data. One of the most popular methods is the Gutenberg–Richter (G–R) law. Also, the knowledge of return period is of great importance in studying and analyzing earthquake hazard and/or seismicity. Such kind of reliable studies strongly depend on (i) selecting a studied area suitable with seismotectonics, (ii) preparing a homogenous catalogue including a large number of earthquakes, and (iii) making a comprehensive completeness analysis and selecting a suitable cut off magnitude, which should be done for each region separately if one studies several seismic zones whose seismotectonics characteristic are different, (iv) selecting a method suitable with data. The accurate estimation of a and b parameters of G–R is of primary importance because the evaluations of the seismicity depend on them. One of the most recent studies on statistical earthquake occurrence can be found in the paper of Sayil and Osmansahin (2008). In their study, Sayil and Osmansahin (2008) used G–R magnitude–frequency relation to investigate the seismicity of western Anatolia. They computed the recurrence periods for the 13 sub-regions of the limited area between 36 –40 N and 26 –32 E. Seismic source zones or sub-regions of an area are widely used to determine the earthquake hazard and earthquake risk of any area in the world. Seismic source zones are determined with two fundamental tools: a seismicity profile and the tectonic structure of the region under consideration (Erdik et al. 1999). Since Turkey is located in a very active seismic area, several studies have investigated the seismic hazard and seismic risk considering different seismic zones in and around Turkey (Alptekin 1978; Yaltirak et al. 1998; Erdik et al. 1985, 1999; Papaioannou and Papazachos 2000; Jimenez et al. 2001; Kayabali 2002; Karavos et al. 2003; Bayrak et al. 2005). Although the authors in all of these studies or other authors studying seismic hazard in different areas of the world use polygonal or rectangular seismic zones according to fault geometries and space distribution of earthquakes, Sayil and Osmansahin (2008) preferred to use elliptical seismic zones. I think that their subregions are not more realistic for studying seismicity of western Anatolia because the