BULLETIN OF THE SEISMOLOGICAL SOCIETY OF AMERICA, cilt.106, sa.3, ss.956-970, 2016 (SCI-Expanded)
Published results of coda Q show a large variation in values. These variations are often claimed to be related to different tectonics, whereas they might just be related to using different assumptions in the processing, leading to different input parameters for the analysis. In this study, the effect of using different processing parameters is investigated and significant differences, particularly at low frequencies, are observed. We find a new set of optimal parameters, which we recommend using in future studies. Using a short lapse time of 30 s and optimal parameters, data from both similar and very different tectonic regions are used to calculate coda Q using the same program and the same parameters. The regions considered are eastern Anatolia, the Azores, Jan Mayen, northwestern and central Argentina, the Shanxi rift system in China, and southwestern Norway. We obtain the following relations: eastern Anatolia (Q = 88 f(0.66)), Azores (Q = 86 f(0.70)), Jan Mayen (Q = 90 f(0.72)), northwestern and central Argentina (Q = 89 f(0.94)), Shanxi rift system (Q = 99 f(0.89)), and southwestern Norway (Q = 124 f(0.91)).