DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación
Artículo científico (JournalArticle)
Broadband Empirical Green’s Function Extraction With Data Adaptive Phase Correlations
17 pages, 11 figures, 2 appendixes.-- The data used in this study were obtained from the Incorporated Research Institutions of Seismology (IRIS) and the European Integrated Data Archive (EIDA). The CPU and GPU accelerated versions of code implementing CC1b,PCC and WPCC, and the two-stage ts-PWS are open source under the LGPL v3 license and available at https://github.com/sergiventosa/FastPCC, and at https://github.com/sergiventosa/ ts-PWS.-- © 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, Seismic ambient noise has a strongly nonstationary time-frequency statistics that demand better interstation correlation methods to improve the balance of highly variable ambient-noise sources, reduce the influence of high-energy poorly-distributed signals, and accelerate the convergence to a robust signal or a broadband empirical Green’s function (EGF). The wavelet cross-correlation method is convenient to analyze the statistics of the cross-correlation of two signals in terms of lag time and scale/frequency. Here, we introduce wavelet phase cross-correlation (WPCC) functions better adapted to the statistics of seismic ambient noise by combining the ideas of: using only the instantaneous phase information of the phase cross-correlation (PCC) method to assess amplitude unbiasedness, and analyzing cross-correlations scale by scale of the wavelet cross-correlation to balance strong and weak frequency components. These features allow WPCC to extract clean broadband signals useful for seismic imaging and monitoring studies. Further, we analyze and discuss benefits and limitations of WPCC compared to PCC in two examples using low-frequency seismic ambient noise (hum), but results are easily extrapolated to higher frequencies. In hum autocorrelations, WPCC can correct for the baseline and extract cleaner signals, and in hum cross-correlations, can extract richer and more frequency-balanced EGFs allowing for a significant increase of Rayleigh phase- and group-velocity measurements and an improvement of their accuracy. Finally, we offset the increase in computational cost of WPCC compared to PCC by using the graphics processing unit (GPU), and show that WPCC is an efficient approach that permits processing large data volumes as commonly encountered in seismic interferometry studies, This work was supported by the SANIMS Project under Grant RTI2018-095594-B-I00. S. Ventosa was also supported in part by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 and in part by the European Union Next Generation (EU/PRTR) program through the PSI Project under Grant PLEC2021-007875. M. Schimmel was also supported by European Union’s Horizon Europe research and innovation programme through the AGEMERA project under Grant 10.3030/101058178. ICM has had funding support of the “Severo Ochoa Centre of Excellence” accreditation (CEX2019-000928-S) of the Spanish Research Agency, Peer reviewed