Coseismic and postseismic elastic wave velocity variations caused by the 2008 Iwate-Miyagi Nairiku earthquake, Japan

• Published in: Journal of Geophysical Research: Solid Earth Vol. 117; no. B9
• Main Authors: Hobiger, M., Wegler, U., Shiomi, K., Nakahara, H.
• Format: Journal Article
• Language: English
• Published: Washington, DC Blackwell Publishing Ltd 01.09.2012; American Geophysical Union
• Subjects: cross correlation; Earth sciences; Earth, ocean, space; Earthquakes; Elastic waves; Exact sciences and technology; Geophysics; Interferometry; Plate tectonics; Seasonal variations; Seismic activity; seismic coda; seismic noise; Seismology; Sensors; Surface waves; temporal variation; Tomography; Velocity; Wave velocity; Far East; Asia; Japan; algorithms; models; cross-correlation; interferometry; Time series; S-waves; elastic waves; recovery; earthquakes; Green function; Long term; frequency; depth; tomography; noise; fault zones; Wave velocity; correlation function
• ISSN: 0148-0227; 2169-9313; 2156-2202; 2169-9356
• DOI: 10.1029/2012JB009402

We analyze coseismic and postseismic velocity variations caused by the June 13, 2008 Iwate‐Miyagi Nairiku earthquake (MW= 6.9) using Passive Image Interferometry (PII). Seismic noise is correlated in order to reconstruct the Green's function between two sensors. Shear wave velocity changes are determined by relating the coda parts of the daily Green's functions to a long‐term reference Green's function. Our study extends from January 2008 to August 2010 and includes the correlations for 190 station pairs in three different frequency ranges from 0.125 to 1.0 Hz. We show that combining the 9 different component cross‐correlation functions stabilizes the velocity change estimation and increases analysis resolution. The observed velocity change curves can be fitted by model time series consisting of a coseismic velocity drop followed by logarithmic postseismic recovery and seasonal velocity variations. The coseismic velocity drops are stronger at higher frequencies and are concentrated in the southern part of the fault zone. A tomography algorithm was developed to reproject the observed velocity variations of the different sensor pairs onto the single sensors. The depth distribution of coseismic changes was modeled for the three stations with the largest velocity drops. At two stations, the coseismic velocity changes are located in the upper several hundred meters. The third station shows indications for deeper changes, in the order of kilometers. Postseismic recovery takes significantly longer than the analyzed two year period. Seasonal velocity variations with periods of one year are observed at all analyzed frequencies for most station pairs. Key Points Resolution increase by merging all 9 component combinations of cross‐correlation Simple tomography of velocity variations Observation of co‐ and postseismic velocity and seasonal variations