Rayleigh-Wave H/V via Noise Cross Correlation in Southern California
- Creators
-
Muir, Jack B.
-
Tsai, Victor C.
Abstract
We study the crustal structure of southern California by inverting horizontal‐to‐vertical (H/V) amplitudes of Rayleigh waves observed in noise cross‐correlation signals. This study constitutes a useful addition to traditional phase‐velocity‐based tomographic inversions due to the localized sensitivity of H/V measurements to the near surface of the measurement station site. The continuous data of 222 permanent broadband stations of the Southern California Seismic Network (SCSN) were used in production of noise cross‐correlation waveforms, resulting in a spatially dense set of measurements for the southern California region in the 1–15 s period band. The fine interstation spacing of the SCSN allows retrieval of high signal‐to‐noise ratio Rayleigh waves at periods as low as 1 s, significantly improving the vertical resolution of the resulting tomographic image, compared to previous studies with minimum periods of 5–10 s. In addition, horizontal resolution is naturally improved by increased station density. Tectonic subregions including the Los Angeles basin and Salton trough are clearly visible due to their high short‐period H/V ratios, whereas the Transverse and Peninsular Ranges exhibit low H/V at all periods.
Additional Information
© 2017 Seismological Society of America. Manuscript received 13 February 2017; Published Online 25 September 2017. The authors would like to acknowledge Daniel Bowden for his thoughts and advice during initial code development. Jack Muir would also like to acknowledge the General Sir John Monash Foundation and the Origin Energy Foundation for their financial support. We would also like to thank the three anonymous reviewers and Associate Editor Eric Chael for their input to the article.Errata
The authors would like to correct an error in the article by Muir and Tsai (2017). In the original article, a programming error resulted in the reported periods being incorrectly divided by 2.5, which is the Nyquist frequency of the data (5 Hz sample rate). Thus, to correct the results, all periods reported for data should be multiplied by 2.5; the period range of the article is consequently 2.5–37.5 s. This error does not change the qualitative conclusions of the article—including the observation that the horizontal‐to‐vertical (H/V) data predict shallower effective basin depth—but does alter the quantitative conclusions of the article because the periods in the original article are 2.5 times too small. Note that the 10 s reported for Figure 1 is correct because this was a synthetic calculation. The change in periods necessitates recalculating the inversion results of Figures 4 and 5 and ? Figure S51 of the supplemental content; the corrected figures are supplied. In addition, for ease of use, we have supplied ? Tables S1, S2, and S3 with relabeled columns with the corrected periods and a corrected table caption noting that the standard deviations are of log H/V; these tables are otherwise unchanged. Acknowledgments: The authors would like to thank Ellen Syracuse for identifying the problem and its source within the codes used to process data. The authors would also like to thank Editor‐in‐Chief Thomas Pratt and Betty Schiefelbein for reviewing this erratum.Attached Files
Supplemental Material - 2017051_esupp.zip
Supplemental Material - 2019187_esupp.zip
Files
| Name | Size | Download all |
|---|---|---|
|
md5:d535c1dcc7d3abf0dd9d9d8ff7438a36
|
31.2 MB | Preview Download |
|
md5:7105aa89907e8bcd37daca942fde6a4f
|
184.6 kB | Preview Download |
Additional details
- Eprint ID
- 82873
- Resolver ID
- CaltechAUTHORS:20171102-082548374
- General Sir John Monash Foundation
- Origin Energy Foundation
- Created
-
2017-11-02Created from EPrint's datestamp field
- Updated
-
2023-06-01Created from EPrint's last_modified field
- Caltech groups
- Seismological Laboratory