Seismic noise analysis in the microseismic and high-frequency domain at the BURAR seismic stations

Abstract

"In the absence of earthquakes, seismic stations continuously record the Earth's vibrations, called ambient seismic noise (ASN). The main concern regarding the ASN records is improving the seismic data quality using different tools developed especially for this purpose. Power Spectral Densities (PSD) and their corresponding Probability Density Functions (PDF) are tools used to evaluate the station performance and reveal the noise level at the station’s site. For high frequencies (>1 Hz), the noise sources are from cultural activities and show diurnal variation, while for low frequencies (<1 Hz), the noise is generated by natural sources and shows seasonal variation. In this study, we analyzed the vertical components of the Bucovina array seismic network (BURAR) to characterize the noise levels at the stations and investigate noise variations in space and time. We computed four-year spectrograms for some elements of the array that revealed an increase of ASN in the 2–5 Hz frequency range during the warmer seasons, contrary to the colder months. On the other hand, at lower frequencies (0.5–1 Hz), the power of seismic noise increases during the year’s colder months. We examined the relationship between noise levels and weather parameters (e.g., wind speed) for the station where the seismic sensor is collocated, with a weather station. We observed that an increase in wind speed leads to an increase in the noise level at high frequencies (> 2 Hz)."