Passive surface wave imaging with ultrashort noise records in urban areas

Abstract

Passive surface-wave methods are widely used in urban areas because of their economic efficiency. It usually needs long-time observation to provide phase velocity information at frequencies ranging from several Hz to several tens of Hz. A complex urbanized environment, however, leads to a reduction in acquisition time. Consequently, it’s necessary to extract surface waves with ultrashort traffic noise for engineering applications. Although noise segments are limited in the circumstance of ultrashort acquisition, we still need data selection to avoid stacking neutralization of data with a different signal-to-noise ratio (SNR). Then we compare linear stacking, phase-weighted stacking (PWS), and time-frequency phase-weighted stacking (tf-PWS) through processing one-minute noise records after selection. We can obtain clear and continuous dispersion energy with these three methods. Our results indicate that PWS or tf-PWS has better performance than linear stacking by comparing dispersion energy images, and PWS improves SNR most, strongly contrasting before and after selection.