Evidence of Dynamic Crustal Deformation in Tohoku, Japan, From Time‐Varying Receiver Functions

Abstract

Temporal variation of crustal structure is key to our understanding of Earth processes on human timescales. Often, we expect that the most significant structural variations are caused by strong ground shaking associated with large earthquakes, and recent studies seem to confirm this. Here we test the possibility of using P receiver functions (PRF) to isolate structural variations over time. Synthetic receiver function tests indicate that structural variation could produce PRF changes on the same order of magnitude as random noise or contamination by local earthquakes. Nonetheless, we find significant variability in observed receiver functions over time at several stations located in northeastern Honshu. Immediately following the Tohoku‐oki earthquake, we observe high PRF variation clustering spatially, especially in two regions near the beginning and end of the rupture plane. Due to the depth sensitivity of PRF and the timescales over which this variability is observed, we infer this effect is primarily due to fluid migration in volcanic regions and shear stress/strength reorganization. While the noise levels in PRF are high for this type of analysis, by sampling small data sets, the computational cost is lower than other methods, such as ambient noise, thereby making PRF a useful tool for estimating temporal variations in crustal structure.

Seismic evidence for megathrust fault-valve behavior during episodic tremor and slip

“…biased by the teleseismic source distribution (31,40). Furthermore, RFs may contain substantial noise that is difficult to quantify.…”

“…Temporal stacking RF data sample a medium repeatedly beneath a seismic station; however, temporal sampling is highly irregular, and the results can be biased by the teleseismic source distribution (31,40). Furthermore, RFs may contain substantial noise that is difficult to quantify.…”

Groundwater Variations From Autocorrelation and Receiver Functions

“…In order to monitor such active subsurface process using seismology, one needs a reproducible seismic signal propagating through the system. Controlled artificial sources have been successfully used for this purpose (e.g., Niu et al, 2008) especially in the petroleum production context (e.g., Silver et al, 2007), but small changes in subsurface properties due to external perturbations can also be monitored using seismic energy generated by earthquakes or ambient noise (e.g., Audet, 2010;Larose et al, 2015;Meier et al, 2010;Porritt & Yoshioka, 2017;Snieder & Page, 2007). Time-lapse monitoring can be performed across a range of timescales, from short ones associated with high precipitation events, through seasonal variations, to long-term changes due to groundwater extraction and drought.…”

Azimuthal Seismic Anisotropy in the Upper Crust of the Japanese Islands Induced by the 2011 Tohoku‐Oki Megathrust Earthquake

“…Takada & Fukushima, 2013). Lithospheric displacements caused by the megathrust earthquake perturbed the regional crust, incorporating static stress field changes (Hong et al, 2015;Porritt & Yoshioka, 2017). The perturbations of the medium were particularly strong in thermally weakened regions (Takada & Fukushima, 2013).…”

“…Strong seismic waves accompanied large dynamic stress changes, which were particularly apparent in volcanic regions with incorporated surface subsidence and seismic velocity reduction (Brenguier et al, ; Takada & Fukushima, ). Lithospheric displacements caused by the megathrust earthquake perturbed the regional crust, incorporating static stress field changes (Hong et al, ; Porritt & Yoshioka, ). The perturbations of the medium were particularly strong in thermally weakened regions (Takada & Fukushima, ).…”