Monitoring temporal variations in instrument responses in regional broadband seismic network using ambient seismic noise

ABSTRACT High‐quality broadband data are required to promote the development of seismology research. Instrument response errors that affect data quality are often difficult to detect from visual waveform inspection alone. Here, we propose a method that uses ambient noise data in the period range of...

Full description

Saved in:
Bibliographic Details
Published in:Geophysical Prospecting 2018-06, Vol.66 (5), p.1019-1036
Main Authors: Ye, Fang, Lin, Jun, Shi, Zhaomin, Lyu, Shixue
Format: Article
Language:English
Subjects:
Ambient noise
Ambient seismic noise
Amplitude
Broadband
Broadband seismic instrument
Correlation
Data
Data analysis
Data processing
Errors
Inspection
Instrument response
Instrument responses
Mathematical analysis
Methods
Noise
Parameter estimation
Parameters
Pollution monitoring
Seismic engineering
Seismic response
Seismology
Sensitivity analysis
Stations
Surface waves
Temporal variations
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:ABSTRACT High‐quality broadband data are required to promote the development of seismology research. Instrument response errors that affect data quality are often difficult to detect from visual waveform inspection alone. Here, we propose a method that uses ambient noise data in the period range of 5−25 s to monitor instrument performance and check data quality in situ. Amplitude information of coda waves and travel time of surface waves extracted from cross‐correlations of ambient noise are used to assess temporal variations in the sensitivity and poles–zeros of instrument responses. The method is based on an analysis of amplitude and phase index parameters calculated from pairwise cross‐correlations of three stations, which provides multiple references for reliable error estimates. Index parameters calculated daily during a two‐year observation period are evaluated to identify stations with instrument response errors in real time. During data processing, initial instrument responses are used in place of available instrument responses to simulate instrument response errors, which are then used to verify our results. The coda waves of noise cross‐correlations help mitigate the effects of a non‐isotropic field and make the amplitude measurements quite stable. Additionally, effects of instrument response errors that experience pole–zero variations on monitoring temporal variations in crustal properties appear statistically significant of velocity perturbation and larger than the standard deviation. Monitoring seismic instrument performance helps eliminate data pollution before analysis begins.
ISSN:0016-8025
1365-2478
DOI:10.1111/1365-2478.12621