Premium
Evidence of Overtone Interference in Fundamental‐Mode Rayleigh Wave Phase and Amplitude Measurements
Author(s) -
Hariharan Anant,
Dalton Colleen A.,
Ma Zhitu,
Ekström Göran
Publication year - 2020
Publication title -
journal of geophysical research: solid earth
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.983
H-Index - 232
eISSN - 2169-9356
pISSN - 2169-9313
DOI - 10.1029/2019jb018540
Subject(s) - seismogram , interference (communication) , amplitude , phase (matter) , rayleigh wave , noise (video) , physics , rayleigh scattering , arc (geometry) , wavelength , geodesy , optics , geology , computational physics , channel (broadcasting) , seismology , wave propagation , geometry , telecommunications , computer science , mathematics , quantum mechanics , artificial intelligence , image (mathematics)
Abstract We present evidence that measurements of minor‐arc fundamental‐mode (FM) Rayleigh waves experience interference from major‐arc overtones, resulting in travel‐time and amplitude measurements oscillating along a ray path. The oscillations are present in synthetic seismograms generated in a 3‐D Earth model via SPECFEM3D_GLOBE and in a 1‐D Earth model by mode summation. The absence of oscillations in synthetics containing only the FM indicates that the oscillations originate from higher‐mode interference. This interference is present across multiple measurement techniques, including multi‐channel cross correlation, phase‐matched filtering, and cluster analysis. Experiments with 1‐D synthetics suggest that contamination from interference is largest at epicentral distances greater than around 120 degrees, where record sections of seismograms show the major‐arc overtones intersecting the minor‐arc FM Rayleigh wave. The short wavelength of the interference pattern means it is only observable with dense station spacing and high data quality, which may explain why it has not, to our knowledge, been previously recognized. We show the interference is visible in real data. Its overall impact on phase‐velocity maps is probably minor due to many measurements from shorter path lengths less prone to interference bias. However, phase‐velocity maps constructed only from measurements at epicentral distances prone to interference exhibit significant noise and poor agreement with maps from measurements that include all path lengths; the issue is especially problematic for approaches that use differential travel times between nearby stations. Accounting for interference may diminish noise in measurements and improve the accuracy of images of the upper mantle.