Reply to Bayrak’s comments on “An investigation of seismicity for western anatolia” by Sayil and Osmanşahin (2008)

Abstract

I looked over the comments of Mr. Critle, and can say that this is not a scientific ‘‘criticization’’. It is very clear that subject of our paper is one of the general seismicity studies based on statistical methods for a defined region. Besides, it has similar format with such a kind of studies in literature (Papazachos 1989; Papazachos and Papaioannou 1993; Sayil and Osmansahin 2004; Sayil and Osmansahin 2005a, b; Sayil and Osmansahin 2006; Sayil 2007) Mr. Critle says in his first criticism: ‘‘they did not form sub-regions suitable with geometrics of the fault systems in the region’’. We did not study on geometrics or mechanisms of the fault systems. We have used eliptic forms as usual (above references) to show the earthquakes occurred in each sub-region. Other criticisms are about our data. It is very clear that we have used the data of the most known and reliable centres and scientists. These catalogues (given in Tables 1 and 2 in our article, Sayil and Osmansahin 2008) had included intensity (Io), body wave magnitude (mb) and surface wave magnitude (Ms) of earthquakes. A new catalogue with main shocks (Ms C 4.0) has been developed from these catalogues for investigation area in our article. In this study (Sayil and Osmansahin 2008), the standard least square (linear regression) method has been used for obtaining the correlations of Ms-mb and Ms-Io. Standard deviations were found as r = 0.17 and r = 0.14, respectively (Fig. 3 in Sayil and Osmansahin 2008) and this method can be used in these error limits. The linear regression method is the most important for statistical investigation, and it is mainly used in literature (Gutenberg and Richter 1954; Alptekin 1978; Erdik et al. 1985; Erdik et al. 1999; Ambraseys 2001; Panza and Romanelli (2001) ; Kayabali 2002; other references in Sayil and Osmansahin 2008). Computed Ms-Io and Ms-mb relations are consistent with those of Ambraseys (2001) and Bath (1966). Other regression methods (polinomial regression,