Seismic Noise Characteristics at the Romanian Broadband Seismic Network

Abstract

The noise level of the Romanian broadband stations operating since 2006 has been studied in order to identify the variations in background seismic noise as a function of time of day, season, and particular conditions at the stations. Power spectral densities and their corresponding probability density functions are used in this paper to characterize the background seismic noise. At high frequencies (> 1 Hz), seismic noise seems to have cultural origin, since significant variations (up to 55 dB) between daytime and nighttime noise levels are observed at almost all of the stations. However, this variability appears not to influence the detection capabilities of the Romanian Seismic Network in case of intermediate-depth earthquakes and earthquakes with magnitude over 3.0. For smaller magnitude events (Mw < 3), we showed that the number of stations detecting an event decreases during daytime. We studied the seasonal variation of the seismic noise for primary and secondary microseisms (with emphasize for the latter) and we demonstrated that the noise levels are higher in winter than in summer. We also observed a shift of the double-frequency peak from lower periods in summer to longer periods in winter. The polarization analysis indicated that the main sources of secondary microseisms for stations close to the Black Sea are coming from the Black Sea, while for the others the main sources are found in the Mediterranean Sea and Atlantic Ocean. Finally, the analysis of the probability density functions for stations located in different geologic conditions has pointed out that the noise level is higher for stations sited on softer formations than those sited on hard rocks.

Keywords:

• Background Seismic Noise
• Probability Density Functions
• Cultural Noise
• Microseisms
• Diurnal Variations
• Seasonal Variations

Acknowledgments

This work was supported by CNCSIS-UEFISCDI, project number PN II-RU 120/2010. Most of the figures were created using Generic Mapping Tool (GMT) [Wessel and Smith, 1998] software package. Figure 11 was created using PQLX [McNamara and Boaz, 2005] software. We are grateful to the reviewer whose suggestions and critical remarks significantly improved the quality of the article.