Mahsa Safarkhani - M.Sc., Institute of Geophysics, University of Tehran, Tehran, Iran
Taghi Shirzad - Assistant Professor, Institute of Geophysics, Polish Academy of Sciences, ۳۰-۱۲۶, Gabrieli Zapolskiej ۴۴, Krakow, Poland

The ambient seismic noise correlation technique has become a standard method to extract empirical Green s functions when the noise sources distribution are considered uniform around the station pairs. In fact, the distribution of noise sources in the Earth would be not ideal (homogeneous and isotropic). In this study, we considered estimating effect of the noise source distributions on retrieved empirical Green s functions signals. In addition to the signals emitted from nonstationary seismic noise sources, the repetition and energy/amplitude of the stationary emitted signals have a detrimental effect on the reconstruction of empirical Green s functions. In this regard, we present Weighted RMS (Root-Mean-Square) stacking method to alleviate non-uniform energy distribution and normalize the flow of energy which comes from a specific direction. We use two stacking methods (Linear, Weighted RMS) for 13 months ambient seismic noise recorded from December 2011 to 2012 at 7 short period stations deployed in the NW of Iran. Our results indicate that Weighted RMS stacking method tackles problem including low signal to noise ratio value compared to linear one. Furthermore, noise correlation functions selection based on special conditions before and during the stacking procedure is caused to obtain stable and reliable empirical Green s functions using Weighted RMS stacking method

Ambient seismic noise, Distributions of noise sources, Empirical Green s Function, Energy flux, WRMS stacking